R/FMP.R
In MislavSag/findata: Collect data fro trading and financial ML
#' @title FMP Class
#'
#' @description
#' Get data data from FMP Cloud.
#'
#' @export
FMP = R6::R6Class(
"FMP",
inherit = DataAbstract,
public = list(
#' @field api_key API KEY for FMP Cloud
api_key = NULL,
#' @field base_url Base URL for the FMP Cloud API.
base_url = "https://financialmodelingprep.com/api",
#' @description
#' Create a new FMP object.
#'
#' @param api_key API KEY for FMP cloud data.
#' @param context_with_config AWS S3 Tiledb config
#' @param base_url Base URL for the FMP Cloud API.
#'
#' @return A new `FMP` object.
initialize = function(api_key = NULL,
context_with_config = NULL,
base_url = NULL) {
# endpoint
# Set base url
self$base_url = base_url
# check and define variables
if (is.null(api_key)) {
self$api_key = assert_character(Sys.getenv("APIKEY-FMPCLOUD"))
} else {
self$api_key = api_key
}
},
#' @description
#' Create a new FMP object.
#'
#' @param path to save reulst data.table
#' @param start_date First date to scrape from.If NULL, date is 2010-01-01.
#'
#' @return Earning announcements data.
get_earning_announcements = function(path, start_date = NULL) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(arrow)
# library(nanotime)
# self = FMP$new()
# symbol = "CA"
# start_date = NULL
# start_date = as.Date("2023-08-01")
# path = "F:/equity/usa/fundamentals/earning_announcements.parquet"
# help function
get_ea = function(start_date, end_date) {
# define listing dates
dates_from = seq.Date(start_date, end_date, by = 3)
dates_to = dates_from + 14
# get data
ea <- lapply(seq_along(dates_from), function(i) {
private$fmpv_path("earning_calendar", from = dates_from[i], to = dates_to[i], apikey = self$api_key)
})
dt <- rbindlist(ea, fill = TRUE)
# clean data
dt <- unique(dt, by = c("symbol", "date"))
dt[, date := as.Date(date)]
dt[, fiscalDateEnding := as.Date(fiscalDateEnding)]
dt[, updatedFromDate := as.Date(updatedFromDate)]
}
# define start date
if (is.null(start_date)) {
start_date = as.Date("2010-01-01")
dt = get_ea(start_date, Sys.Date())
} else {
dt_existing = arrow::read_parquet(path)
dt_new = get_ea(start_date, Sys.Date())
if (nrow(dt_new) == 0) {
print("No data for earning announcements.")
return(NULL)
} else {
dt = unique(rbind(dt_existing, dt_new), by = c("date", "symbol", "updatedFromDate"))
}
}
# save to uri
arrow::write_parquet(dt, path)
},
#' @description Get earnings suprises data from FMP cloud Prep.
#'
#' @param year Year.
#'
#' @return data.table with earnings surprises data.
get_earnings_suprises = function(year = format(Sys.Date(), "%Y")) {
url = "https://financialmodelingprep.com/api/v4/earnings-surprises-bulk"
p = RETRY("GET", url, query = list(apikey = self$api_key, year = year))
res = content(p)
setDT(res)
return(res)
},
#' @description Get Earning Call Transcript from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#' @param uri to save transcripts data for every symbol.
#'
#' @return Result of GET request
get_transcripts = function(symbols, uri) {
# debug
# library(data.table)
# library(httr)
# library(findata)
# library(arrow)
# library(lubridate)
# self = list()
# self$api_key = Sys.getenv("APIKEY-FMPCLOUD")
# # get symbols for transcripts
# events = arrow::read_parquet("F:/equity/usa/fundamentals/earning_announcements.parquet")
# events = as.data.table(events)
# events <- events[date < Sys.Date()] # remove announcements for today
# events <- na.omit(events, cols = c("eps")) # remove rows with NA for earnings
# url <- modify_url("https://financialmodelingprep.com/", path = "api/v3/available-traded/list",
# query = list(apikey = Sys.getenv("APIKEY-FMPCLOUD")))
# res <- GET(url)
# stocks <- rbindlist(content(res), fill = TRUE)
# usa_symbols <- stocks[exchangeShortName %in% c("AMEX", "NASDAQ", "NYSE", "OTC")]
# events <- events[symbol %in% usa_symbols$symbo]
# symbols <- unique(events$symbol)
# uri = "F:/equity/usa/fundamentals/transcripts"
# main loop
url_meta <- "https://financialmodelingprep.com/api/v4/earning_call_transcript"
lapply(symbols, function(s) {
# debug
print(s)
# get transcripts metadata for symbol
p <- GET(url_meta, query = list(symbol = s, apikey = self$api_key))
res <- rbindlist(content(p))
if (nrow(res) > 0) {
res = cbind.data.frame(symbol = s, res)
} else {
return(NULL)
}
setnames(res, c("symbol", "quarter", "year", "date"))
# create dir
file_ = file.path(uri, paste0(s, ".parquet"))
if (file.exists(file_)) {
old_data = arrow::read_parquet(file_)
new = fsetdiff(as.data.table(res[, c("quarter", "year")]),
as.data.table(old_data[, c("quarter", "year")]))
years = new[, unique(year)]
if (length(years) == 0) {
return(NULL)
}
} else {
years = unique(res$year)
}
# get new data
new_data <- lapply(years, function(y) {
private$fmpv_path(paste0("batch_earning_call_transcript/", s), v = "v4",
year = y, apikey = self$api_key)
})
new_data <- rbindlist(new_data)
if (nrow(new_data) > 0) {
new_data$date <- as.POSIXct(new_data$date, tz = "America/New_York")
} else {
print("There are no transcripts data.")
}
# clean data
new_data[, date := with_tz(date, tzone = "UTC")]
new_data <- na.omit(new_data)
new_data <- unique(new_data, by = c("symbol", "date"))
# rbind old and new data
if (file.exists(file_)) {
new_data = unique(rbind(old_data, new_data))
}
# save
arrow::write_parquet(new_data, file_)
return(NULL)
})
},
#' @description get daily data from FMP cloud for all stocks
#'
#' @param uri to save daily data for every date.
#'
#' @return Scrap all daily data
get_daily_tiledb = function(uri) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(arrow)
# library(lubridate)
# self = FMP$new()
# uri = "F:/equity/daily_fmp"
# create data seq
seq_date_all <- getBusinessDays(as.Date("1990-01-01"), Sys.Date() - 1)
# # read old data
files = list.files(uri)
if (length(files) == 0) {
seq_date = seq_date_all
} else {
seq_date = as.Date(setdiff(as.character(seq_date_all), tools::file_path_sans_ext(files)))
}
# get daily with batch and save to azure blob
url_base <- "https://financialmodelingprep.com/api/v4/batch-request-end-of-day-prices"
lapply(seq_date, function(x) {
print(x)
file_name_ = file.path(uri, paste0(x, ".csv"))
p <- RETRY("GET",
url_base,
query = list(date = x, apikey = self$api_key),
write_disk(file_name_),
times = 5L)
return(NULL)
})
},
#' @description retrieving intraday market data from FMP cloud.
#'
#' @param symbol Symbol of the stock.
#' @param multiply multiplier.Start date.
#' @param time Size of the time. (minute, hour, day, week, month, quarter, year)
#' @param from Start date,
#' @param to End date.
#'
#' @return Data frame with ohlcv data.
get_intraday_equities = function(symbol,
multiply = 1,
time = 'day',
from = as.character(Sys.Date() - 3),
to = as.character(Sys.Date())) {
# initial GET request. Don't use RETRY here yet.
x <- tryCatch({
GET(paste0('https://financialmodelingprep.com/api/v4/historical-price/',
symbol, '/', multiply, '/', time, '/', from, '/', to),
query = list(apikey = self$api_key),
timeout(100))
}, error = function(e) NA)
# control error
tries <- 0
while (all(is.na(x)) & tries < 20) {
print("There is an error in scraping market data. Sleep and try again!")
Sys.sleep(60L)
x <- tryCatch({
GET(paste0('https://financialmodelingprep.com/api/v4/historical-price/',
symbol, '/', multiply, '/', time, '/', from, '/', to),
query = list(apikey = self$api_key),
timeout(100))
}, error = function(e) NA)
tries <- tries + 1
}
# check if status is ok. If not, try to download again
if (x$status_code == 404) {
print("There is an 404 error!")
return(NULL)
} else if (x$status_code == 200) {
x <- content(x)
return(rbindlist(x$results))
} else {
x <- RETRY("GET",
paste0('https://financialmodelingprep.com/api/v4/historical-price/',
symbol, '/', multiply, '/', time, '/', from, '/', to),
query = list(apikey = self$api_key),
times = 5,
timeout(100))
if (x$status_code == 200) {
x <- content(x)
return(rbindlist(x$results))
} else {
stop('Error in reposne. Status not 200 and not 404')
}
}
},
#' #' @description Get tick quote data from FMP cloud.
#' #'
#' #' @param symbol Symbol of the stock.
#' #' @param date multiplier.Start date.
#' #'
#' #' @return Data frame with ohlcv data.
#' get_quotes = function(symbol, date) {
#'
#' # debug
#' # library(httr)
#' # library(data.table)
#' # self = list()
#' # self$api_key = Sys.getenv("APIKEY-FMPCLOUD")
#' # symbol = "SPY"
#' # date = "2023-08-18"
#'
#' # init list
#' q = list()
#'
#' # initial GET request. Don't use RETRY here yet.
#' x <- tryCatch({
#' GET(paste0('https://financialmodelingprep.com/api/v4/historical-price-tick/',
#' symbol, '/', date),
#' query = list(limit = 30000, apikey = self$api_key),
#' timeout(100))
#' }, error = function(e) NA)
#' x <- content(x)
#' q[[1]] = rbindlist(x$results)
#'
#' # get last time
#' last_time = tail(q[[1]]$t, 1)
#'
#' x <- tryCatch({
#' GET(paste0('https://financialmodelingprep.com/api/v4/historical-price-tick/',
#' symbol, '/', date),
#' query = list(limit = 300, ts = last_time, te = last_time + 1000, apikey = self$api_key),
#' timeout(100))
#' }, error = function(e) NA)
#' x <- content(x)
#' rbindlist(x$results)
#' q[[2]] = rbindlist(x$results)
#'
#'
#'
#'
#' # url = "https://financialmodelingprep.com/api/v4/historical-price-tick/SPY/2023-08-04?limit=500&apikey=15cd5d0adf4bc6805a724b4417bbaafc"
#' # GET(url)
#'
#' # control error
#' tries <- 0
#' while (all(is.na(x)) & tries < 20) {
#' print("There is an error in scraping market data. Sleep and try again!")
#' Sys.sleep(60L)
#' x <- tryCatch({
#' GET(paste0('https://financialmodelingprep.com/api/v4/historical-price/',
#' symbol, '/', multiply, '/', time, '/', from, '/', to),
#' query = list(apikey = self$api_key),
#' timeout(100))
#' }, error = function(e) NA)
#' tries <- tries + 1
#' }
#'
#' # check if status is ok. If not, try to download again
#' if (x$status_code == 404) {
#' print("There is an 404 error!")
#' return(NULL)
#' } else if (x$status_code == 200) {
#' x <- content(x)
#' return(rbindlist(x$results))
#' } else {
#' x <- RETRY("GET",
#' paste0('https://financialmodelingprep.com/api/v4/historical-price/',
#' symbol, '/', multiply, '/', time, '/', from, '/', to),
#' query = list(apikey = self$api_key),
#' times = 5,
#' timeout(100))
#' if (x$status_code == 200) {
#' x <- content(x)
#' return(rbindlist(x$results))
#' } else {
#' stop('Error in reposne. Status not 200 and not 404')
#' }
#' }
#' },
#' @description Help function for calculating start and end dates
#'
#' @param start_dates Sequence of start dates.
#' @param n number of consecutive dates.
#'
#' @return list of start and end dates
create_start_end_dates = function(start_dates, n = 4) {
# rearange date sequences
last_date <- start_dates[1]
start_dates_ <- c(last_date)
for (s in start_dates) {
if ((as.integer(s) - as.integer(last_date)) >= n) { # num of trading days to scrap
start_dates_ <- c(start_dates_, s)
last_date <- s
} else {
next()
}
}
start_dates_ <- as.Date(start_dates_, origin = "1970-01-01")
end_dates <- c(start_dates_[-1], tail(start_dates, 1) + (n + 1))
end_dates <- end_dates - 1
if (tail(end_dates, 1) > (Sys.Date() - 1)) {
end_dates[length(end_dates)] <- Sys.Date() - 1
}
return(list(start_dates = start_dates_, end_dates = end_dates))
},
#' @description Get SP500 symbols from fmp cloud.
#'
#' @return Vector of SP500 symbols.
get_sp500_symbols = function() {
# get SP 500 stocks from FMP LCOUD
SP500 <- content(GET(paste0('https://financialmodelingprep.com/api/v3/sp500_constituent?apikey=', self$api_key)))
SP500 <- rbindlist(SP500)
SP500_DELISTED <- GET(paste0('https://financialmodelingprep.com/api/v3/historical/sp500_constituent?apikey=', self$api_key))
SP500_DELISTED <- rbindlist(content(SP500_DELISTED))
SP500_SYMBOLS <- unique(c(SP500$symbol, SP500_DELISTED$symbol))
# get symbol changes
utils_changes <- UtilsData$new()
SP500_CHANGES <- lapply(SP500_SYMBOLS, utils_changes$get_ticker_data)
SP500_CHANGES <- rbindlist(SP500_CHANGES)
SP500_SYMBOLS <- unique(c(SP500_SYMBOLS, SP500_CHANGES$ticker_change))
return(SP500_SYMBOLS)
},
#' @description Create factor files for calculating adjusted prices (adjusted for splits and dividends).
#'
#' @param save_uri Uri where tiledb saves files
#' @param prices_uri Uri to import prices from.
#'
#' @return Factor files saved to "factor-file" blob.
get_factor_files = function(save_uri = "s3://equity-usa-factor-files",
prices_uri = "s3://equity-usa-daily-fmpcloud") {
# DEBUG
# library(httr)
# library(findata)
# library(data.table)
# library(findata)
# library(tiledb)
# library(lubridate)
# library(rvest)
# self = FMP$new()
# save_uri = "s3://equity-usa-factor-files"
# prices_uri = "s3://equity-usa-daily-fmpcloud"
# DEBUG
# TODO use dividends from stocksalysis andcheck aIG then. Dividend 3$ wrong
# read daily unadjusted daily data
arr <- tiledb_array(prices_uri,
as.data.frame = TRUE,
query_layout = "UNORDERED")
system.time(unadj_daily_data <- arr[])
# delete object
del_obj <- tryCatch(tiledb_object_rm(save_uri), error = function(e) NA)
if (is.na(del_obj)) {
warning("Can't delete object")
}
# order by symbol and date
DT <- as.data.table(unadj_daily_data)
DT <- unique(DT, by = c("symbol", "date"))
setorder(DT, symbol, date)
symbols <- unique(DT$symbol)
DT <- DT[, .(symbol, date , close)]
# splits
utils <- UtilsData$new()
splits_l <- lapply(symbols, utils$get_split_factor)
names(splits_l) <- symbols
splits <- rbindlist(splits_l, fill = TRUE, idcol = "symbol")
df <- splits[DT, on = c("symbol", "date")]
df[, ratio := NULL]
# dividends
url <- "https://financialmodelingprep.com/api/v3/historical-price-full/stock_dividend/"
dividends_l <- lapply(symbols, function(x) {
y <- RETRY("GET", paste0(url, x), query = list(apikey = self$api_key), times = 5L)
y <- content(y)
y <- rbindlist(y$historical, fill = TRUE)
y <- cbind(symbol = x, y)
return(y)
})
dividends <- rbindlist(dividends_l, fill = TRUE)
dividends[, date := as.Date(date)]
dividends <- dividends[date > as.Date("2004-01-01")]
dividends[is.na(dividend), diviend := adjDividend]
dividends <- dividends[, .(symbol, date, dividend)]
df <- dividends[df, on = c("symbol", "date")]
# create factor file
df[, `:=`(lag_close = shift(close), date = shift(date)), by = symbol]
df <- df[!is.na(date)]
factor_files <- df[(!is.na(dividend) & dividend != 0) | !is.na(split_factor),
.(date, dividend, split_factor, lag_close), by = symbol]
# create empty factor files for equities with no dividends and stock splits
empty_factor_file_symbols <- setdiff(symbols, unique(factor_files$symbol))
df_empty <- data.table(date = c(as.Date("2004-01-02"),
as.Date("2050-01-02")),
price_factor = c(1, 1),
split_factor = c(1, 1),
lag_close = c(1, 0))
factor_empty <- merge(data.frame(symbol = empty_factor_file_symbols),
df_empty)
factor_empty <- as.data.table(factor_empty)
setorder(factor_empty, symbol, date)
# create facotr files for othe equities
### DEBUG #######
# df <- splits[DT, on = c("symbol", "date")]
# df[, ratio := NULL]
# df <- dividends[df, on = c("symbol", "date")]
# df[, `:=`(lag_close = shift(close), date = shift(date)), by = symbol]
# df <- df[!is.na(date)]
# factor_files <- df[(!is.na(dividend) & dividend != 0) | !is.na(split_factor),
# .(date, dividend, split_factor, lag_close), by = symbol]
# factor_files <- factor_files[date %between% c("2004-01-01", "2021-04-01")]
### DEBUG #######
# cont creating factor files
factor_files <- unique(factor_files)
factor_files[, split_factor := nafill(split_factor, type = "nocb"), by = symbol]
factor_files[is.na(split_factor), split_factor := as.numeric(1), by = symbol]
end_row_by_group <- data.table(symbol = unique(factor_files$symbol),
date = as.Date("2050-01-01"),
dividend = NA,
split_factor = 1,
lag_close = 0)
factor_files <- rbind(factor_files, end_row_by_group)
setorder(factor_files, symbol, date)
create_price_factor <- function(lag_close, split_factor, dividend) {
price_factor <- vector("numeric", length(lag_close))
for (i in length(lag_close):1) {
if (i == length(lag_close)) {
price_factor[i] <- (lag_close[i - 1] - dividend[i -
1])/lag_close[i - 1] * 1
}
else if (i == 1) {
price_factor[i] <- NA
}
else if (is.na(dividend[i - 1])) {
price_factor[i] <- price_factor[i + 1]
}
else {
price_factor[i] <- ((lag_close[i - 1] - dividend[i -
1])/lag_close[i - 1]) * price_factor[i + 1]
}
}
price_factor <- c(price_factor[-1], NA)
price_factor[is.na(price_factor)] <- 1
return(price_factor)
}
factor_files[, price_factor := create_price_factor(lag_close, split_factor, dividend),
by = symbol]
factor_files <- factor_files[, .(symbol, date, price_factor, split_factor,
lag_close)]
# debug
# factor_files[symbol == "ABT"]
# add empty
factor_files_final <- rbind(factor_empty, factor_files)
setorder(factor_files_final, symbol, date)
# save factor file to blob
if (tiledb_object_type(save_uri) != "ARRAY") {
fromDataFrame(
obj = as.data.frame(factor_files_final),
uri = save_uri,
col_index = c("symbol", "date"),
sparse = TRUE,
allows_dups = TRUE
)
}
return(factor_files)
},
#' @description Get IPO date from fmp cloud
#'
#' @param ticker Stock ticker
#'
#' @return Stock IPO date.
get_ipo_date = function(ticker) {
url <- "https://financialmodelingprep.com/api/v4/company-outlook"
p <- content(GET(url, query = list(symbol = ticker, apikey = self$api_key)))
if ("error" %in% names(p)) {
return("2004-01-01")
} else {
return(p$profile$ipoDate)
}
},
#' @description Get ipo-calendar-confirmed date from fmp cloud
#'
#' @param start_date Start date
#' @param end_date End date.
#'
#' @param uri TileDB uri argument
#'
#' @return Stock IPO date.
get_ipo_calendar_confirmed_bulk = function(start_date,
end_date = Sys.Date()) {
# start_date = Sys.Date() - 19
# Meta
url <- "https://financialmodelingprep.com/api/v4/ipo-calendar-confirmed"
if ((end_date - start_date) > 20) {
seq_date_ <- unique(c(seq.Date(start_date, end_date, by = 20), end_date))
start_date <- seq_date_[1:(length(seq_date_)-1)]
end_date <- seq_date_[-1]
}
# Get ipo data for every data span
ipos <- lapply(seq_along(start_date), function(i) {
content(GET(url, query = list(from = start_date[i],
to = end_date[i],
apikey = self$api_key)))
})
# Merge and clean data
ipo_data <- lapply(ipos, rbindlist)
ipo_data <- rbindlist(ipo_data)
ipo_data[, `:=`(filingDate = as.Date(filingDate),
acceptedDate = as.POSIXct(acceptedDate, tz = "America/New_York"),
effectivenessDate = as.Date(effectivenessDate))]
ipo_data[, acceptedDate := with_tz(acceptedDate, tzone = "UTC")]
ipo_data <- unique(ipo_data)
return(ipo_data)
},
#' @description Get FI data.
#'
#' @param symbol Stock symbol.
#' @param statement Stock statement.
#' @param period Stock period.
#'
#' @return data.table of financial reports.
get_fi_statement = function(symbol,
statement = c("income-statement", "balance-sheet-statement", "cash-flow-statement"),
period = c("annual", "quarter")) {
url <- "https://financialmodelingprep.com/api/v3"
url <- paste(url, statement, symbol, sep = "/")
p <- RETRY("GET", url, query = list(period = period, apikey = self$api_key, limit = 10000))
res <- content(p)
res <- rbindlist(res)
return(res)
},
#' @description Get financial ratios, company key metrics.
#'
#' @param symbol Stock symbol.
#' @param type Type of fundamental analysis.
#' @param period Stock period.
#'
#' @return data.table of financial reports.
get_financial_metrics = function(symbol,
type = c("ratios", "key-metrics", "financial-growth"),
period = c("annual", "quarter")) {
url <- "https://financialmodelingprep.com/api/v3"
url <- paste(url, type, symbol, sep = "/")
p <- RETRY("GET", url, query = list(period = period, apikey = self$api_key, limit = 10000))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get stock splits data from FMP cloud Prep.
#'
#' @param start_date Start date
#' @param end_date End date.
#'
#' @return data.table of stock splits
get_stock_splits = function(start_date = Sys.Date() - 5,
end_date = Sys.Date()) {
url <- "https://financialmodelingprep.com/api/v3/stock_split_calendar"
p <- RETRY("GET", url, query = list(from = start_date,
to = end_date,
apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get market cap data from FMP cloud Prep.
#'
#' @param ticker Ticker.
#' @param limit Limit.
#'
#' @return data.table with market cap data.
get_market_cap = function(ticker, limit = 10) {
url <- "https://financialmodelingprep.com/api/v3/historical-market-capitalization/"
url <- paste0(url, toupper(ticker))
p <- RETRY("GET", url, query = list(limit = limit, apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get market cap data from FMP cloud Prep - bulk.
#'
#' @param symbols stock symbols.
#' @param uri to market cap data.
#'
#' @return data.table with market cap data.
get_market_cap_bulk = function(symbols, uri) {
# main loop
lapply(symbols, function(s) {
# debug
# library(findata)
# library(arrow)
# library(data.table)
# self = FMP$new()
# uri = "F:/equity/usa/fundamentals/market_capitalization"
# [1] "CDXC"
# [1] "Get new data"
# [1] "KATE"
# [1] "Get new data"
# Error in as.Date.default(date) :
# do not know how to convert 'date' to class “Date”
# In addition: Warning message:
# In `[.data.table`(data_new, , `:=`(symbol, NULL)) :
# Column 'symbol' does not exist to remove
print(s)
# file name
file_name = file.path(uri, paste0(s, ".parquet"))
# get all existing data
if (file.exists(file_name)) {
dt_old = arrow::read_parquet(file_name)
limit = as.integer(Sys.Date() - dt_old[, max(as.Date(date))])
if (limit == 0) return(NULL)
} else {
limit = 10000
}
# get new data
print("Get new data")
data_new <- mapply(
self$get_market_cap,
ticker = s,
limit = limit,
SIMPLIFY = FALSE
)
data_new = data_new[[1]]
if (length(data_new) == 0) {
return(NULL)
}
data_new[, symbol := NULL]
data_new[, date := as.Date(date)]
# merge if necessary
if (file.exists(file_name)) {
data_new = unique(rbind(dt_old, data_new))
}
setorder(data_new, -date)
# save
arrow::write_parquet(data_new, file_name)
return(NULL)
})
},
#' @description Get stock list.
#'
#' @return data.table with stock list data.
get_stock_list = function() {
url <- "https://financialmodelingprep.com/api/v3/stock/list"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get symbol changes.
#'
#' @return data.table with symbol changes data.
get_symbol_changes = function() {
url <- "https://financialmodelingprep.com/api/v4/symbol_change"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get delisted companies.
#'
#' @return data.table with delisted companies data.
get_delisted_companies = function() {
url <- "https://financialmodelingprep.com/api/v3/delisted-companies"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get available traded list.
#'
#' @return data.table with available traded list data.
get_available_traded_list = function() {
url <- "https://financialmodelingprep.com/api/v3/available-traded/list"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get SP500 stocks.
#'
#' @return data.table with SP500 tocks.
get_sp500_constituent = function() {
url <- "https://financialmodelingprep.com/api/v3/sp500_constituent"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get Financial Statements data
#'
#' @param uri uri path to the directory.
#' @param years Get statement for specific year.
#' @param statement quarter or annual.
#' @param period quarter or annual.
#'
#' @return data.table of financial reports.
get_fi_statement_bulk = function(uri,
years = 1999:(data.table::year(Sys.Date())),
statement = c("income-statement-bulk",
"balance-sheet-statement-bulk",
"cash-flow-statement-bulk",
"ratios-bulk",
"key-metrics-bulk",
"financial-growth-bulk",
"income-statement-growth-bulk",
"balance-sheet-statement-growth-bulk",
"cash-flow-statement-growth-bulk"),
period = c("annual", "quarter")
) {
# DEBUG
# library(findata)
# library(httr)
# library(data.table)
# uri = "F:/equity/usa/fundamentals"
# years = 1990:(data.table::year(Sys.Date()))
# statement ="rating-bulk"
# period = "quarter"
# self <- FMP$new()
# define save_uri
save_uri = file.path(uri, statement, period)
if (!dir.exists(save_uri)) {
dir.create(save_uri, recursive = TRUE)
}
# prepare donwload data
url <- "https://financialmodelingprep.com/api/v4"
url <- paste(url, statement, sep = "/")
# get new years to download
years_exist = gsub(".csv", "", list.files(save_uri))
years = c(setdiff(years, years_exist), year(Sys.Date()))
# donwload data
lapply(years, function(year) {
print(year)
file_name = file.path(save_uri, paste0(year, ".csv"))
RETRY("GET",
url,
query = list(year = year, period = period, apikey = self$api_key),
write_disk(file_name, overwrite = TRUE))
return(NULL)
})
},
#' @description Help function for getting grades and targets
#'
#' @param symbols stock symbols.
#' @param tag url tag.
#'
#' @return Result of GET request
get_daily_v4 = function(symbols, tag = "price-target") {
# get data for tagfor all symbols
dt_l <- lapply(symbols, function(s) {
private$fmpv_path(
path = tag,
v = "v4",
symbol = s,
apikey = self$api_key
)
})
dt <- rbindlist(dt_l)
# clean data
dt <- na.omit(dt, c("symbol"))
dt <- unique(dt)
return(dt)
},
#' @description Get Upgrades and Downgrades from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#'
#' @return Result of GET request
get_grades = function(symbols) {
# get all grades for all symbols
grades = self$get_daily_v4(symbols, "upgrades-downgrades")
grades[, publishedDate := as.POSIXct(publishedDate,
format = "%Y-%m-%dT%H:%M:%OSZ",
tz = "UTC")]
return(grades)
},
#' @description Get Price Targest from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#'
#' @return Result of GET request
get_targets = function(symbols) {
# get all targets for all symbols
targets = self$get_daily_v4(symbols, "price-target")
targets[, publishedDate := as.POSIXct(publishedDate,
format = "%Y-%m-%dT%H:%M:%OSZ",
tz = "UTC")]
return(targets)
},
#' @description Get Ratings from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#' @param uri uri to save output data.
#' @param update Should we scrape all data od jut the new one.
#' @param update_limit how many days to retrieve if update eqtual to TRUE.
#'
#' @return Result of GET request
get_ratings = function(symbols,
uri = "ratings.parquet",
update = FALSE,
update_limit = 10) {
# debug
# library(findata)
# library(httr)
# library(data.table)
# library(arrow)
# self = FMP$new()
# uri = "F:/equity/usa/fundamentals/ratings.parquet"
# private = list(
# ea_file_name = "EarningAnnouncements",
# transcripts_file_name = "earnings-calendar.rds",
# prices_file_name = "prices.csv",
# market_cap_file_name = "MarketCap",
#
# fmpv_path = function(path = "earning_calendar", v = "v3", ...) {
#
# # query params
# query_params <- list(...)
#
# # define url
# url <- paste0("https://financialmodelingprep.com/api/", v, "/", path)
#
# # get data
# p <- RETRY("GET", url, query = query_params, times = 5)
# result <- suppressWarnings(rbindlist(httr::content(p), fill = TRUE))
#
# return(result)
# }
# )
# get symbols for transcripts
# events = arrow::read_parquet("F:/equity/usa/fundamentals/earning_announcements.parquet")
# events = as.data.table(events)
# events <- events[date < Sys.Date()] # remove announcements for today
# events <- na.omit(events, cols = c("eps")) # remove rows with NA for earnings
# url <- modify_url("https://financialmodelingprep.com/", path = "api/v3/available-traded/list",
# query = list(apikey = Sys.getenv("APIKEY-FMPCLOUD")))
# res <- GET(url)
# stocks <- rbindlist(content(res), fill = TRUE)
# usa_symbols <- stocks[exchangeShortName %in% c("AMEX", "NASDAQ", "NYSE", "OTC")]
# events <- events[symbol %in% usa_symbols$symbo]
# symbols <- unique(events$symbol)
# help function
ratings_post = function(dt) {
ratings <- rbindlist(ratings_l)
ratings[, date := as.Date(date)]
# keep only numeric columns, without description
cols_ <- c("symbol", "date", "rating",
colnames(ratings)[grep("Score", colnames(ratings))])
ratings <- ratings[, ..cols_]
}
# get new data if update is FALSE
if (update == FALSE) {
# get all ratings for all symbols
ratings_l <- lapply(symbols, function(s) {
private$fmpv_path(
path = paste0("historical-rating/", s),
v = "v3",
limit = 10000,
apikey = self$api_key
)
})
ratings = ratings_post(ratings_l)
} else {
# import existing data
ratings_raw <- as.data.table(arrow::read_parquet(uri))
# get new data
ratings_l <- lapply(symbols, function(s) {
private$fmpv_path(
path = paste0("historical-rating/", s),
v = "v3",
limit = update_limit,
apikey = self$api_key
)
})
ratings = ratings_post(ratings_l)
# rbind
ratings = rbind(ratings_raw, ratings)
}
# clean data
ratings <- na.omit(ratings, c("symbol", "date"))
ratings <- unique(ratings)
# save to uri
arrow::write_parquet(ratings, uri)
return(NULL)
},
#' @description Get Historical dividends from FMP cloud.
#'
#' @param symbols Vector of stock symbols.
#'
#' @return Data table with historical dividends for symbol.
get_historical_dividends = function(symbols) {
# dividends
urls <- paste0("https://financialmodelingprep.com/api/v3/historical-price-full/stock_dividend/",
symbols)
dividends_l <- lapply(urls, function(x) {
p <- RETRY("GET", x, query = list(apikey = self$api_key))
res <- httr::content(p)
div <- rbindlist(res$historical)
cbind(symbol = res$symbol, div)
})
dividends <- rbindlist(dividends_l, fill = TRUE)
return(dividends)
},
#' @description Get hour data for all history from fmp cloud and save as parquet on AWS.
#'
#' @param symbols Symbol of the stock.
#' @param uri_minute AWS S3 bucket uri or NAS path.
#' @param save_uri_hour AWS S3 bucket uri for hour data.
#' @param save_uri_daily AWS S3 bucket uri for daily data.
#' @param deep_scan should we test for dates with low number od observation
#' and try to scrap again.
#'
#' @return Data saved to uri.
get_equities_batch = function(symbols,
uri_minute = "s3://equity-usa-minute",
deep_scan = FALSE) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(RcppQuantuccia)
# library(lubridate)
# library(nanotime)
# library(arrow)
# library(checkmate)
# library(future.apply)
# self = FMP$new()
# symbols = c("AAPL", "SPY")
# uri_minute = "F:/equity/usa/minute" # "s3://equity-usa-minute"
# deep_scan = FALSE
# fmp = FMP$new()
# library(rvest)
# sp100 <- read_html("https://en.wikipedia.org/wiki/S%26P_100") |>
# html_elements(x = _, "table") |>
# (`[[`)(3) |>
# html_table(x = _, fill = TRUE)
# sp100_symbols <- c("SPY", "TLT", "GLD", "USO", "SVXY",
# sp100$Symbol)
# symbols <- fmp$get_sp500_symbols()
# symbols <- na.omit(symbols)
# symbols <- symbols[!grepl("/", symbols)]
# symbols <- unique(union(sp100_symbols, symbols))
# check if we have all necessary env variables
assert_choice("AWS_ACCESS_KEY_ID", names(Sys.getenv()))
assert_choice("AWS_SECRET_ACCESS_KEY", names(Sys.getenv()))
assert_choice("AWS_DEFAULT_REGION", names(Sys.getenv()))
# # s3 bucket
# if (grepl("^s3:/", uri_minute)) {
# bucket = s3_bucket(uri_minute)
# dir_files = bucket$ls()
# } else {
# dir_files = list.files(uri_minute)
# }
# parallel execution
# plan("multisession", workers = 2L)
# ADD FUTURE LAPPLY BELOW
# main loop to scrap minute data
lapply(symbols, function(s) {
# debug
print(s)
# define file name and s3 file name
file_name = paste0(s, ".parquet")
file_name_full = file.path(uri_minute, file_name)
# define start_dates
start_dates <- seq.Date(as.Date("2004-01-01"), Sys.Date() - 1, by = 1)
start_dates <- start_dates[isBusinessDay(start_dates)]
# get trading days from daily data
print("Get daily data")
daily_start <- seq.Date(as.Date("2004-01-01"), Sys.Date(), by = 365 * 4)
daily_end <- c(daily_start[-1], Sys.Date())
daily_data <- lapply(seq_along(daily_start), function(i) {
daily_data <- tryCatch({
self$get_intraday_equities(s,
multiply = 1,
time = 'day',
from = daily_start[i],
to = daily_end[i])
}, error = function(e) NULL)
})
daily_data <- rbindlist(daily_data)
if (length(daily_data) == 0) return(NULL)
start_dates <- as.Date(intersect(start_dates, as.Date(daily_data$formated)), origin = "1970-01-01")
# TODO: check BRK-b (BTK.B) and other symbols with -/.
# OLD DATA ----------------------------------------------------------------
# read old data
if (file_name %in% dir_files) {
# read data for the symbol
print("Get minute data")
data_history = arrow::read_parquet(file_name_full)
# cont if there is history data
if (length(data_history) > 0) {
# basic clean
data_history <- unique(data_history, by = "date")
setorder(data_history, date)
# missing freq
if (deep_scan) { # CHECK THIS LATER !!!
# create date column
data_history_tz <- copy(data_history)
# data_history_tz[, time := as.POSIXct(time, origin = as.POSIXct("1970-01-01 00:00:00", tz = "UTC"), tz = "UTC")]
# data_history_tz[, time := with_tz(time, tz = "America/New_York")]
# data_history_tz[, time_ := as.Date(time, tz = "America/New_York")]
# data_history_tz[, time_n := .N, by = time_]
#
# # define dates we will try to scrap again
# observation_per_day <- 60 * 6
# try_again <- unique(data_history_tz[time_n < observation_per_day, time_])
# start_dates <- as.Date(intersect(try_again, start_dates), origin = "1970-01-01")
# end_dates <- start_dates
} else {
# define dates to scrap
data_history_date = data_history[, unique(as.Date(date, tz = "America/New_York"))]
# get final dates to scrap
start_dates <- as.Date(setdiff(start_dates, data_history_date), origin = "1970-01-01")
end_dates <- start_dates
# data_history[as.Date(date) == as.Date("2004-01-16")]
}
} else {
dates <- self$create_start_end_dates(start_dates, 1)
start_dates <- dates$start_dates
end_dates <- dates$end_dates
}
} else {
dates <- self$create_start_end_dates(start_dates, 1)
start_dates <- dates$start_dates
end_dates <- dates$end_dates
}
# GET NEW DATA ------------------------------------------------------------
# if there is no dates next
if (length(start_dates) == 0) {
print(paste0("No data for symbol ", s))
return(NULL)
}
# get data
print("Get new minute data")
system.time({
data_slice = lapply(seq_along(start_dates), function(d) {
# for (d in seq_along(start_dates[1:10])) {
if (end_dates[d] >= Sys.Date()) end_dates[d] <- Sys.Date() - 1
self$get_intraday_equities(
s,
multiply = 1,
time = "minute",
from = start_dates[d],
to = end_dates[d]
)
})
})
if (any(unlist(sapply(data_slice, nrow)) > 4999)) {
stop("More than 4999 rows!")
}
data_by_symbol <- rbindlist(data_slice)
# if there is no data next
if (nrow(data_by_symbol) == 0) {
print(paste0("No data for symbol ", s))
return(NULL)
}
# convert to numeric (not sure why I put these, but it have sense I believe).
cols <- c("o", "h", "l", "c", "v")
data_by_symbol[, (cols) := lapply(.SD, as.numeric), .SDcols = cols]
# clean data
data_by_symbol <- unique(data_by_symbol, by = c("formated"))
data_by_symbol[, date := as.POSIXct(t / 1000,
origin = as.POSIXct("1970-01-01 00:00:00", tz = "UTC"),
tz = "UTC")]
data_by_symbol = data_by_symbol[, .(date, o, h, l, c, v)]
setnames(data_by_symbol, c("date", "open", "high", "low", "close", "volume"))
# rbind new an old
if (file_name %in% dir_files && nrow(data_history) > 0) {
# rbind new data
data_by_symbol = rbind(data_history, data_by_symbol)
setorder(data_by_symbol, "date")
data_by_symbol = unique(data_by_symbol, by = "date")
# delete old parquet file
if (grepl("^s3:/", uri_minute)) {
bucket$DeleteFile(file_name)
} else {
file.remove(file_name_full)
}
# save file
arrow::write_parquet(data_by_symbol, file_name_full)
} else {
# save to S3
arrow::write_parquet(data_by_symbol, file_name_full)
}
return(NULL)
})
},
#' @description Get minute data for all history from FMP cloud.
#'
#' @param symbols Symbol of the stock.
#' @param uri_minute AWS S3 bucket uri or NAS path.
#' @param deep_scan should we test for dates with low number od observation
#' and try to scrap again.
#' @param workers Number of workers.
#'
#' @return Data saved to uri.
get_minute = function(symbols,
uri_minute,
deep_scan = FALSE,
workers = 1L) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(RcppQuantuccia)
# library(lubridate)
# library(nanotime)
# library(arrow)
# library(checkmate)
# library(future.apply)
# self = FMP$new()
# symbols = c("AAPL", "SPY")
# uri_minute = "F:/equity/usa/minute"
# deep_scan = FALSE
# fmp = FMP$new()
# library(rvest)
# sp100 <- read_html("https://en.wikipedia.org/wiki/S%26P_100") |>
# html_elements(x = _, "table") |>
# (`[[`)(3) |>
# html_table(x = _, fill = TRUE)
# sp100_symbols <- c("SPY", "TLT", "GLD", "USO", "SVXY",
# sp100$Symbol)
# symbols <- fmp$get_sp500_symbols()
# symbols <- na.omit(symbols)
# symbols <- symbols[!grepl("/", symbols)]
# symbols <- unique(union(sp100_symbols, symbols))
# parallel execution
# plan("multisession", workers = 2L)
# ADD FUTURE LAPPLY BELOW
# main loop to scrap minute data
mclapply(symbols, function(s) {
# debug
# s = "WBA"
print(s)
# create folder for symbol if it doesn't exists
dir_name = file.path(uri_minute, s)
if (!dir.exists(dir_name)) {
dir.create(dir_name)
}
# get trading days from daily data
print("Get daily data")
daily_start <- seq.Date(as.Date("2004-01-01"), Sys.Date(), by = 365 * 4)
daily_end <- c(daily_start[-1], Sys.Date())
daily_data <- lapply(seq_along(daily_start), function(i) {
daily_data <- tryCatch({
self$get_intraday_equities(s,
multiply = 1,
time = 'day',
from = daily_start[i],
to = daily_end[i])
}, error = function(e) NULL)
})
daily_data <- rbindlist(daily_data)
if (length(daily_data) == 0) return(NULL)
dates <- as.Date(daily_data$formated)
# keep only NYSE business days
dates_business = getBusinessDays(min(dates), max(dates))[]
dates_business = dates_business[dates_business %in% dates]
dates = dates_business
# extract missing dates
dates_exists = gsub(".parquet", "", list.files(dir_name))
dates_new = as.Date(setdiff(paste0(dates), dates_exists))
# get new data
for (i in seq_along(dates_new)) {
# crate date
date = dates_new[i]
print(date)
# get new data for date
date_data = self$get_intraday_equities(
s,
multiply = 1,
time = "minute",
from = date,
to = date
)
# if there is no data next
if (nrow(date_data) == 0) {
print(paste0("No data for date ", date, " for symbol ", s))
date_data = data.table(
o = NA_real_,
h = NA_real_,
l = NA_real_,
c = NA_real_,
v = NA_real_,
t = NA_real_,
formated = "no minute data"
)
}
# save file
file_name_full = file.path(dir_name, paste0(date, ".parquet"))
arrow::write_parquet(date_data, file_name_full)
}
}, mc.cores = workers)
},
#' @description Get dividend data from FMP cloud Prep.
#'
#' @param ticker Ticker.
#'
#' @return data.table with dividend data.
get_dividends = function(ticker) {
url <- "https://financialmodelingprep.com/api/v3/historical-price-full/stock_dividend/"
url <- paste0(url, toupper(ticker))
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res$historical, fill = TRUE)
return(res)
},
#' @description Get beneficial ownership from FMP cloud.
#'
#' @param ticker Ticker.
#'
#' @return data.table with beneficial ownership data.
beneficial_ownership = function(ticker) {
# ticker = "AAPL"
# self$api_key <- Sys.getenv("APIKEY-FMPCLOUD")
url <- "https://financialmodelingprep.com/api/v4/insider/ownership/acquisition_of_beneficial_ownership"
p <- RETRY("GET", url, query = list(symbol = toupper(ticker), apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
# QUOTE -------------------------------------------------------------------
#' @description Get the latest bid and ask prices for a stock, as well as the
#' volume and last trade price in real-time.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the full quote for the specified stock.
get_full_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/quote/", symbol)
p = RETRY("GET", url, query = list(apikey = self$api_key), times = 5)
res = content(p)
res = rbindlist(res, fill = TRUE)
res[, time := as.POSIXct(timestamp, origin = "1970-01-01", tz = "America/New_York")]
return(res)
},
#' @description Get a simplified quote for a stock, including the current price, volume, and last trade price.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the quote order for the specified stock.
get_quote_order = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/quote-order/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get a simple quote for a stock, including the price, change, and volume.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the simple quote for the specified stock.
get_simple_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0("https://financialmodelingprep.com/api/v3/quote-short/", symbol)
p = RETRY("GET", url, query = list(apikey = self$api_key), times = 5)
res = content(p)
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get the latest bid and ask prices for an OTC stock, as well as the volume and last trade price in real-time.
#'
#' @param symbol The OTC stock symbol(s) (e.g., "BATRB,FWONB").
#'
#' @return A data.table containing the OTC quote for the specified stock(s).
get_otc_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/otc/real-time-price/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get real-time prices for all stocks listed on a specific exchange.
#'
#' @param exchange The exchange for which to retrieve prices (e.g., "NYSE").
#'
#' @return A data.table containing real-time prices for all stocks on the specified exchange.
get_exchange_prices = function(exchange = "NYSE") {
assert_character(exchange, len = 1, null.ok = FALSE)
url = paste0("https://financialmodelingprep.com/api/v3/quotes/", exchange)
p = RETRY("GET", url, query = list(apikey = self$api_key), times = 5)
res = content(p)
res = rbindlist(res, fill = TRUE)
res[, time := as.POSIXct(timestamp, tz = "America/New_York")]
return(res)
},
#' @description Get the change in a stock's price over a given period of time.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the stock price change for the specified stock.
get_stock_price_change = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/stock-price-change/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get information on trades that have occurred in the aftermarket.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing aftermarket trade information for the specified stock.
get_aftermarket_trades = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/pre-post-market-trade/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = as.data.table(res)
return(res)
},
#' @description Get the latest bid and ask prices for a stock in the aftermarket.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing aftermarket quote information for the specified stock.
get_aftermarket_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/pre-post-market/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = as.data.table(res)
return(res)
},
#' @description Get quotes for multiple stocks at once.
#'
#' @param symbols A comma-separated string of stock symbols (e.g., "AAPL,MSFT").
#'
#' @return A data.table containing batch quote information for the specified stocks.
get_batch_quote = function(symbols) {
assert_character(symbols, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/batch-pre-post-market/", symbols)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get trades for multiple stocks at once.
#'
#' @param symbols A comma-separated string of stock symbols (e.g., "AAPL,MSFT").
#'
#' @return A data.table containing batch trade information for the specified stocks.
get_batch_trade = function(symbols) {
assert_character(symbols, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/batch-pre-post-market-trade/", symbols)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
}
),
private = list(
# PRIVATE -----------------------------------------------------------------
ea_file_name = "EarningAnnouncements",
transcripts_file_name = "earnings-calendar.rds",
prices_file_name = "prices.csv",
market_cap_file_name = "MarketCap",
fmpv_path = function(path = "earning_calendar", v = "v3", ...) {
# query params
query_params <- list(...)
# define url
url <- paste0("https://financialmodelingprep.com/api/", v, "/", path)
# get data
p <- RETRY("GET", url, query = query_params, times = 5)
result <- suppressWarnings(rbindlist(httr::content(p), fill = TRUE))
return(result)
}
)
)
MislavSag/findata documentation built on June 2, 2025, 12:34 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
#' @title FMP Class
#'
#' @description
#' Get data data from FMP Cloud.
#'
#' @export
FMP = R6::R6Class(
"FMP",
inherit = DataAbstract,
public = list(
#' @field api_key API KEY for FMP Cloud
api_key = NULL,
#' @field base_url Base URL for the FMP Cloud API.
base_url = "https://financialmodelingprep.com/api",
#' @description
#' Create a new FMP object.
#'
#' @param api_key API KEY for FMP cloud data.
#' @param context_with_config AWS S3 Tiledb config
#' @param base_url Base URL for the FMP Cloud API.
#'
#' @return A new `FMP` object.
initialize = function(api_key = NULL,
context_with_config = NULL,
base_url = NULL) {
# endpoint
# Set base url
self$base_url = base_url
# check and define variables
if (is.null(api_key)) {
self$api_key = assert_character(Sys.getenv("APIKEY-FMPCLOUD"))
} else {
self$api_key = api_key
}
},
#' @description
#' Create a new FMP object.
#'
#' @param path to save reulst data.table
#' @param start_date First date to scrape from.If NULL, date is 2010-01-01.
#'
#' @return Earning announcements data.
get_earning_announcements = function(path, start_date = NULL) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(arrow)
# library(nanotime)
# self = FMP$new()
# symbol = "CA"
# start_date = NULL
# start_date = as.Date("2023-08-01")
# path = "F:/equity/usa/fundamentals/earning_announcements.parquet"
# help function
get_ea = function(start_date, end_date) {
# define listing dates
dates_from = seq.Date(start_date, end_date, by = 3)
dates_to = dates_from + 14
# get data
ea <- lapply(seq_along(dates_from), function(i) {
private$fmpv_path("earning_calendar", from = dates_from[i], to = dates_to[i], apikey = self$api_key)
})
dt <- rbindlist(ea, fill = TRUE)
# clean data
dt <- unique(dt, by = c("symbol", "date"))
dt[, date := as.Date(date)]
dt[, fiscalDateEnding := as.Date(fiscalDateEnding)]
dt[, updatedFromDate := as.Date(updatedFromDate)]
}
# define start date
if (is.null(start_date)) {
start_date = as.Date("2010-01-01")
dt = get_ea(start_date, Sys.Date())
} else {
dt_existing = arrow::read_parquet(path)
dt_new = get_ea(start_date, Sys.Date())
if (nrow(dt_new) == 0) {
print("No data for earning announcements.")
return(NULL)
} else {
dt = unique(rbind(dt_existing, dt_new), by = c("date", "symbol", "updatedFromDate"))
}
}
# save to uri
arrow::write_parquet(dt, path)
},
#' @description Get earnings suprises data from FMP cloud Prep.
#'
#' @param year Year.
#'
#' @return data.table with earnings surprises data.
get_earnings_suprises = function(year = format(Sys.Date(), "%Y")) {
url = "https://financialmodelingprep.com/api/v4/earnings-surprises-bulk"
p = RETRY("GET", url, query = list(apikey = self$api_key, year = year))
res = content(p)
setDT(res)
return(res)
},
#' @description Get Earning Call Transcript from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#' @param uri to save transcripts data for every symbol.
#'
#' @return Result of GET request
get_transcripts = function(symbols, uri) {
# debug
# library(data.table)
# library(httr)
# library(findata)
# library(arrow)
# library(lubridate)
# self = list()
# self$api_key = Sys.getenv("APIKEY-FMPCLOUD")
# # get symbols for transcripts
# events = arrow::read_parquet("F:/equity/usa/fundamentals/earning_announcements.parquet")
# events = as.data.table(events)
# events <- events[date < Sys.Date()] # remove announcements for today
# events <- na.omit(events, cols = c("eps")) # remove rows with NA for earnings
# url <- modify_url("https://financialmodelingprep.com/", path = "api/v3/available-traded/list",
# query = list(apikey = Sys.getenv("APIKEY-FMPCLOUD")))
# res <- GET(url)
# stocks <- rbindlist(content(res), fill = TRUE)
# usa_symbols <- stocks[exchangeShortName %in% c("AMEX", "NASDAQ", "NYSE", "OTC")]
# events <- events[symbol %in% usa_symbols$symbo]
# symbols <- unique(events$symbol)
# uri = "F:/equity/usa/fundamentals/transcripts"
# main loop
url_meta <- "https://financialmodelingprep.com/api/v4/earning_call_transcript"
lapply(symbols, function(s) {
# debug
print(s)
# get transcripts metadata for symbol
p <- GET(url_meta, query = list(symbol = s, apikey = self$api_key))
res <- rbindlist(content(p))
if (nrow(res) > 0) {
res = cbind.data.frame(symbol = s, res)
} else {
return(NULL)
}
setnames(res, c("symbol", "quarter", "year", "date"))
# create dir
file_ = file.path(uri, paste0(s, ".parquet"))
if (file.exists(file_)) {
old_data = arrow::read_parquet(file_)
new = fsetdiff(as.data.table(res[, c("quarter", "year")]),
as.data.table(old_data[, c("quarter", "year")]))
years = new[, unique(year)]
if (length(years) == 0) {
return(NULL)
}
} else {
years = unique(res$year)
}
# get new data
new_data <- lapply(years, function(y) {
private$fmpv_path(paste0("batch_earning_call_transcript/", s), v = "v4",
year = y, apikey = self$api_key)
})
new_data <- rbindlist(new_data)
if (nrow(new_data) > 0) {
new_data$date <- as.POSIXct(new_data$date, tz = "America/New_York")
} else {
print("There are no transcripts data.")
}
# clean data
new_data[, date := with_tz(date, tzone = "UTC")]
new_data <- na.omit(new_data)
new_data <- unique(new_data, by = c("symbol", "date"))
# rbind old and new data
if (file.exists(file_)) {
new_data = unique(rbind(old_data, new_data))
}
# save
arrow::write_parquet(new_data, file_)
return(NULL)
})
},
#' @description get daily data from FMP cloud for all stocks
#'
#' @param uri to save daily data for every date.
#'
#' @return Scrap all daily data
get_daily_tiledb = function(uri) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(arrow)
# library(lubridate)
# self = FMP$new()
# uri = "F:/equity/daily_fmp"
# create data seq
seq_date_all <- getBusinessDays(as.Date("1990-01-01"), Sys.Date() - 1)
# # read old data
files = list.files(uri)
if (length(files) == 0) {
seq_date = seq_date_all
} else {
seq_date = as.Date(setdiff(as.character(seq_date_all), tools::file_path_sans_ext(files)))
}
# get daily with batch and save to azure blob
url_base <- "https://financialmodelingprep.com/api/v4/batch-request-end-of-day-prices"
lapply(seq_date, function(x) {
print(x)
file_name_ = file.path(uri, paste0(x, ".csv"))
p <- RETRY("GET",
url_base,
query = list(date = x, apikey = self$api_key),
write_disk(file_name_),
times = 5L)
return(NULL)
})
},
#' @description retrieving intraday market data from FMP cloud.
#'
#' @param symbol Symbol of the stock.
#' @param multiply multiplier.Start date.
#' @param time Size of the time. (minute, hour, day, week, month, quarter, year)
#' @param from Start date,
#' @param to End date.
#'
#' @return Data frame with ohlcv data.
get_intraday_equities = function(symbol,
multiply = 1,
time = 'day',
from = as.character(Sys.Date() - 3),
to = as.character(Sys.Date())) {
# initial GET request. Don't use RETRY here yet.
x <- tryCatch({
GET(paste0('https://financialmodelingprep.com/api/v4/historical-price/',
symbol, '/', multiply, '/', time, '/', from, '/', to),
query = list(apikey = self$api_key),
timeout(100))
}, error = function(e) NA)
# control error
tries <- 0
while (all(is.na(x)) & tries < 20) {
print("There is an error in scraping market data. Sleep and try again!")
Sys.sleep(60L)
x <- tryCatch({
GET(paste0('https://financialmodelingprep.com/api/v4/historical-price/',
symbol, '/', multiply, '/', time, '/', from, '/', to),
query = list(apikey = self$api_key),
timeout(100))
}, error = function(e) NA)
tries <- tries + 1
}
# check if status is ok. If not, try to download again
if (x$status_code == 404) {
print("There is an 404 error!")
return(NULL)
} else if (x$status_code == 200) {
x <- content(x)
return(rbindlist(x$results))
} else {
x <- RETRY("GET",
paste0('https://financialmodelingprep.com/api/v4/historical-price/',
symbol, '/', multiply, '/', time, '/', from, '/', to),
query = list(apikey = self$api_key),
times = 5,
timeout(100))
if (x$status_code == 200) {
x <- content(x)
return(rbindlist(x$results))
} else {
stop('Error in reposne. Status not 200 and not 404')
}
}
},
#' #' @description Get tick quote data from FMP cloud.
#' #'
#' #' @param symbol Symbol of the stock.
#' #' @param date multiplier.Start date.
#' #'
#' #' @return Data frame with ohlcv data.
#' get_quotes = function(symbol, date) {
#'
#' # debug
#' # library(httr)
#' # library(data.table)
#' # self = list()
#' # self$api_key = Sys.getenv("APIKEY-FMPCLOUD")
#' # symbol = "SPY"
#' # date = "2023-08-18"
#'
#' # init list
#' q = list()
#'
#' # initial GET request. Don't use RETRY here yet.
#' x <- tryCatch({
#' GET(paste0('https://financialmodelingprep.com/api/v4/historical-price-tick/',
#' symbol, '/', date),
#' query = list(limit = 30000, apikey = self$api_key),
#' timeout(100))
#' }, error = function(e) NA)
#' x <- content(x)
#' q[[1]] = rbindlist(x$results)
#'
#' # get last time
#' last_time = tail(q[[1]]$t, 1)
#'
#' x <- tryCatch({
#' GET(paste0('https://financialmodelingprep.com/api/v4/historical-price-tick/',
#' symbol, '/', date),
#' query = list(limit = 300, ts = last_time, te = last_time + 1000, apikey = self$api_key),
#' timeout(100))
#' }, error = function(e) NA)
#' x <- content(x)
#' rbindlist(x$results)
#' q[[2]] = rbindlist(x$results)
#'
#'
#'
#'
#' # url = "https://financialmodelingprep.com/api/v4/historical-price-tick/SPY/2023-08-04?limit=500&apikey=15cd5d0adf4bc6805a724b4417bbaafc"
#' # GET(url)
#'
#' # control error
#' tries <- 0
#' while (all(is.na(x)) & tries < 20) {
#' print("There is an error in scraping market data. Sleep and try again!")
#' Sys.sleep(60L)
#' x <- tryCatch({
#' GET(paste0('https://financialmodelingprep.com/api/v4/historical-price/',
#' symbol, '/', multiply, '/', time, '/', from, '/', to),
#' query = list(apikey = self$api_key),
#' timeout(100))
#' }, error = function(e) NA)
#' tries <- tries + 1
#' }
#'
#' # check if status is ok. If not, try to download again
#' if (x$status_code == 404) {
#' print("There is an 404 error!")
#' return(NULL)
#' } else if (x$status_code == 200) {
#' x <- content(x)
#' return(rbindlist(x$results))
#' } else {
#' x <- RETRY("GET",
#' paste0('https://financialmodelingprep.com/api/v4/historical-price/',
#' symbol, '/', multiply, '/', time, '/', from, '/', to),
#' query = list(apikey = self$api_key),
#' times = 5,
#' timeout(100))
#' if (x$status_code == 200) {
#' x <- content(x)
#' return(rbindlist(x$results))
#' } else {
#' stop('Error in reposne. Status not 200 and not 404')
#' }
#' }
#' },
#' @description Help function for calculating start and end dates
#'
#' @param start_dates Sequence of start dates.
#' @param n number of consecutive dates.
#'
#' @return list of start and end dates
create_start_end_dates = function(start_dates, n = 4) {
# rearange date sequences
last_date <- start_dates[1]
start_dates_ <- c(last_date)
for (s in start_dates) {
if ((as.integer(s) - as.integer(last_date)) >= n) { # num of trading days to scrap
start_dates_ <- c(start_dates_, s)
last_date <- s
} else {
next()
}
}
start_dates_ <- as.Date(start_dates_, origin = "1970-01-01")
end_dates <- c(start_dates_[-1], tail(start_dates, 1) + (n + 1))
end_dates <- end_dates - 1
if (tail(end_dates, 1) > (Sys.Date() - 1)) {
end_dates[length(end_dates)] <- Sys.Date() - 1
}
return(list(start_dates = start_dates_, end_dates = end_dates))
},
#' @description Get SP500 symbols from fmp cloud.
#'
#' @return Vector of SP500 symbols.
get_sp500_symbols = function() {
# get SP 500 stocks from FMP LCOUD
SP500 <- content(GET(paste0('https://financialmodelingprep.com/api/v3/sp500_constituent?apikey=', self$api_key)))
SP500 <- rbindlist(SP500)
SP500_DELISTED <- GET(paste0('https://financialmodelingprep.com/api/v3/historical/sp500_constituent?apikey=', self$api_key))
SP500_DELISTED <- rbindlist(content(SP500_DELISTED))
SP500_SYMBOLS <- unique(c(SP500$symbol, SP500_DELISTED$symbol))
# get symbol changes
utils_changes <- UtilsData$new()
SP500_CHANGES <- lapply(SP500_SYMBOLS, utils_changes$get_ticker_data)
SP500_CHANGES <- rbindlist(SP500_CHANGES)
SP500_SYMBOLS <- unique(c(SP500_SYMBOLS, SP500_CHANGES$ticker_change))
return(SP500_SYMBOLS)
},
#' @description Create factor files for calculating adjusted prices (adjusted for splits and dividends).
#'
#' @param save_uri Uri where tiledb saves files
#' @param prices_uri Uri to import prices from.
#'
#' @return Factor files saved to "factor-file" blob.
get_factor_files = function(save_uri = "s3://equity-usa-factor-files",
prices_uri = "s3://equity-usa-daily-fmpcloud") {
# DEBUG
# library(httr)
# library(findata)
# library(data.table)
# library(findata)
# library(tiledb)
# library(lubridate)
# library(rvest)
# self = FMP$new()
# save_uri = "s3://equity-usa-factor-files"
# prices_uri = "s3://equity-usa-daily-fmpcloud"
# DEBUG
# TODO use dividends from stocksalysis andcheck aIG then. Dividend 3$ wrong
# read daily unadjusted daily data
arr <- tiledb_array(prices_uri,
as.data.frame = TRUE,
query_layout = "UNORDERED")
system.time(unadj_daily_data <- arr[])
# delete object
del_obj <- tryCatch(tiledb_object_rm(save_uri), error = function(e) NA)
if (is.na(del_obj)) {
warning("Can't delete object")
}
# order by symbol and date
DT <- as.data.table(unadj_daily_data)
DT <- unique(DT, by = c("symbol", "date"))
setorder(DT, symbol, date)
symbols <- unique(DT$symbol)
DT <- DT[, .(symbol, date , close)]
# splits
utils <- UtilsData$new()
splits_l <- lapply(symbols, utils$get_split_factor)
names(splits_l) <- symbols
splits <- rbindlist(splits_l, fill = TRUE, idcol = "symbol")
df <- splits[DT, on = c("symbol", "date")]
df[, ratio := NULL]
# dividends
url <- "https://financialmodelingprep.com/api/v3/historical-price-full/stock_dividend/"
dividends_l <- lapply(symbols, function(x) {
y <- RETRY("GET", paste0(url, x), query = list(apikey = self$api_key), times = 5L)
y <- content(y)
y <- rbindlist(y$historical, fill = TRUE)
y <- cbind(symbol = x, y)
return(y)
})
dividends <- rbindlist(dividends_l, fill = TRUE)
dividends[, date := as.Date(date)]
dividends <- dividends[date > as.Date("2004-01-01")]
dividends[is.na(dividend), diviend := adjDividend]
dividends <- dividends[, .(symbol, date, dividend)]
df <- dividends[df, on = c("symbol", "date")]
# create factor file
df[, `:=`(lag_close = shift(close), date = shift(date)), by = symbol]
df <- df[!is.na(date)]
factor_files <- df[(!is.na(dividend) & dividend != 0) | !is.na(split_factor),
.(date, dividend, split_factor, lag_close), by = symbol]
# create empty factor files for equities with no dividends and stock splits
empty_factor_file_symbols <- setdiff(symbols, unique(factor_files$symbol))
df_empty <- data.table(date = c(as.Date("2004-01-02"),
as.Date("2050-01-02")),
price_factor = c(1, 1),
split_factor = c(1, 1),
lag_close = c(1, 0))
factor_empty <- merge(data.frame(symbol = empty_factor_file_symbols),
df_empty)
factor_empty <- as.data.table(factor_empty)
setorder(factor_empty, symbol, date)
# create facotr files for othe equities
### DEBUG #######
# df <- splits[DT, on = c("symbol", "date")]
# df[, ratio := NULL]
# df <- dividends[df, on = c("symbol", "date")]
# df[, `:=`(lag_close = shift(close), date = shift(date)), by = symbol]
# df <- df[!is.na(date)]
# factor_files <- df[(!is.na(dividend) & dividend != 0) | !is.na(split_factor),
# .(date, dividend, split_factor, lag_close), by = symbol]
# factor_files <- factor_files[date %between% c("2004-01-01", "2021-04-01")]
### DEBUG #######
# cont creating factor files
factor_files <- unique(factor_files)
factor_files[, split_factor := nafill(split_factor, type = "nocb"), by = symbol]
factor_files[is.na(split_factor), split_factor := as.numeric(1), by = symbol]
end_row_by_group <- data.table(symbol = unique(factor_files$symbol),
date = as.Date("2050-01-01"),
dividend = NA,
split_factor = 1,
lag_close = 0)
factor_files <- rbind(factor_files, end_row_by_group)
setorder(factor_files, symbol, date)
create_price_factor <- function(lag_close, split_factor, dividend) {
price_factor <- vector("numeric", length(lag_close))
for (i in length(lag_close):1) {
if (i == length(lag_close)) {
price_factor[i] <- (lag_close[i - 1] - dividend[i -
1])/lag_close[i - 1] * 1
}
else if (i == 1) {
price_factor[i] <- NA
}
else if (is.na(dividend[i - 1])) {
price_factor[i] <- price_factor[i + 1]
}
else {
price_factor[i] <- ((lag_close[i - 1] - dividend[i -
1])/lag_close[i - 1]) * price_factor[i + 1]
}
}
price_factor <- c(price_factor[-1], NA)
price_factor[is.na(price_factor)] <- 1
return(price_factor)
}
factor_files[, price_factor := create_price_factor(lag_close, split_factor, dividend),
by = symbol]
factor_files <- factor_files[, .(symbol, date, price_factor, split_factor,
lag_close)]
# debug
# factor_files[symbol == "ABT"]
# add empty
factor_files_final <- rbind(factor_empty, factor_files)
setorder(factor_files_final, symbol, date)
# save factor file to blob
if (tiledb_object_type(save_uri) != "ARRAY") {
fromDataFrame(
obj = as.data.frame(factor_files_final),
uri = save_uri,
col_index = c("symbol", "date"),
sparse = TRUE,
allows_dups = TRUE
)
}
return(factor_files)
},
#' @description Get IPO date from fmp cloud
#'
#' @param ticker Stock ticker
#'
#' @return Stock IPO date.
get_ipo_date = function(ticker) {
url <- "https://financialmodelingprep.com/api/v4/company-outlook"
p <- content(GET(url, query = list(symbol = ticker, apikey = self$api_key)))
if ("error" %in% names(p)) {
return("2004-01-01")
} else {
return(p$profile$ipoDate)
}
},
#' @description Get ipo-calendar-confirmed date from fmp cloud
#'
#' @param start_date Start date
#' @param end_date End date.
#'
#' @param uri TileDB uri argument
#'
#' @return Stock IPO date.
get_ipo_calendar_confirmed_bulk = function(start_date,
end_date = Sys.Date()) {
# start_date = Sys.Date() - 19
# Meta
url <- "https://financialmodelingprep.com/api/v4/ipo-calendar-confirmed"
if ((end_date - start_date) > 20) {
seq_date_ <- unique(c(seq.Date(start_date, end_date, by = 20), end_date))
start_date <- seq_date_[1:(length(seq_date_)-1)]
end_date <- seq_date_[-1]
}
# Get ipo data for every data span
ipos <- lapply(seq_along(start_date), function(i) {
content(GET(url, query = list(from = start_date[i],
to = end_date[i],
apikey = self$api_key)))
})
# Merge and clean data
ipo_data <- lapply(ipos, rbindlist)
ipo_data <- rbindlist(ipo_data)
ipo_data[, `:=`(filingDate = as.Date(filingDate),
acceptedDate = as.POSIXct(acceptedDate, tz = "America/New_York"),
effectivenessDate = as.Date(effectivenessDate))]
ipo_data[, acceptedDate := with_tz(acceptedDate, tzone = "UTC")]
ipo_data <- unique(ipo_data)
return(ipo_data)
},
#' @description Get FI data.
#'
#' @param symbol Stock symbol.
#' @param statement Stock statement.
#' @param period Stock period.
#'
#' @return data.table of financial reports.
get_fi_statement = function(symbol,
statement = c("income-statement", "balance-sheet-statement", "cash-flow-statement"),
period = c("annual", "quarter")) {
url <- "https://financialmodelingprep.com/api/v3"
url <- paste(url, statement, symbol, sep = "/")
p <- RETRY("GET", url, query = list(period = period, apikey = self$api_key, limit = 10000))
res <- content(p)
res <- rbindlist(res)
return(res)
},
#' @description Get financial ratios, company key metrics.
#'
#' @param symbol Stock symbol.
#' @param type Type of fundamental analysis.
#' @param period Stock period.
#'
#' @return data.table of financial reports.
get_financial_metrics = function(symbol,
type = c("ratios", "key-metrics", "financial-growth"),
period = c("annual", "quarter")) {
url <- "https://financialmodelingprep.com/api/v3"
url <- paste(url, type, symbol, sep = "/")
p <- RETRY("GET", url, query = list(period = period, apikey = self$api_key, limit = 10000))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get stock splits data from FMP cloud Prep.
#'
#' @param start_date Start date
#' @param end_date End date.
#'
#' @return data.table of stock splits
get_stock_splits = function(start_date = Sys.Date() - 5,
end_date = Sys.Date()) {
url <- "https://financialmodelingprep.com/api/v3/stock_split_calendar"
p <- RETRY("GET", url, query = list(from = start_date,
to = end_date,
apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get market cap data from FMP cloud Prep.
#'
#' @param ticker Ticker.
#' @param limit Limit.
#'
#' @return data.table with market cap data.
get_market_cap = function(ticker, limit = 10) {
url <- "https://financialmodelingprep.com/api/v3/historical-market-capitalization/"
url <- paste0(url, toupper(ticker))
p <- RETRY("GET", url, query = list(limit = limit, apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get market cap data from FMP cloud Prep - bulk.
#'
#' @param symbols stock symbols.
#' @param uri to market cap data.
#'
#' @return data.table with market cap data.
get_market_cap_bulk = function(symbols, uri) {
# main loop
lapply(symbols, function(s) {
# debug
# library(findata)
# library(arrow)
# library(data.table)
# self = FMP$new()
# uri = "F:/equity/usa/fundamentals/market_capitalization"
# [1] "CDXC"
# [1] "Get new data"
# [1] "KATE"
# [1] "Get new data"
# Error in as.Date.default(date) :
# do not know how to convert 'date' to class “Date”
# In addition: Warning message:
# In `[.data.table`(data_new, , `:=`(symbol, NULL)) :
# Column 'symbol' does not exist to remove
print(s)
# file name
file_name = file.path(uri, paste0(s, ".parquet"))
# get all existing data
if (file.exists(file_name)) {
dt_old = arrow::read_parquet(file_name)
limit = as.integer(Sys.Date() - dt_old[, max(as.Date(date))])
if (limit == 0) return(NULL)
} else {
limit = 10000
}
# get new data
print("Get new data")
data_new <- mapply(
self$get_market_cap,
ticker = s,
limit = limit,
SIMPLIFY = FALSE
)
data_new = data_new[[1]]
if (length(data_new) == 0) {
return(NULL)
}
data_new[, symbol := NULL]
data_new[, date := as.Date(date)]
# merge if necessary
if (file.exists(file_name)) {
data_new = unique(rbind(dt_old, data_new))
}
setorder(data_new, -date)
# save
arrow::write_parquet(data_new, file_name)
return(NULL)
})
},
#' @description Get stock list.
#'
#' @return data.table with stock list data.
get_stock_list = function() {
url <- "https://financialmodelingprep.com/api/v3/stock/list"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get symbol changes.
#'
#' @return data.table with symbol changes data.
get_symbol_changes = function() {
url <- "https://financialmodelingprep.com/api/v4/symbol_change"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get delisted companies.
#'
#' @return data.table with delisted companies data.
get_delisted_companies = function() {
url <- "https://financialmodelingprep.com/api/v3/delisted-companies"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get available traded list.
#'
#' @return data.table with available traded list data.
get_available_traded_list = function() {
url <- "https://financialmodelingprep.com/api/v3/available-traded/list"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get SP500 stocks.
#'
#' @return data.table with SP500 tocks.
get_sp500_constituent = function() {
url <- "https://financialmodelingprep.com/api/v3/sp500_constituent"
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get Financial Statements data
#'
#' @param uri uri path to the directory.
#' @param years Get statement for specific year.
#' @param statement quarter or annual.
#' @param period quarter or annual.
#'
#' @return data.table of financial reports.
get_fi_statement_bulk = function(uri,
years = 1999:(data.table::year(Sys.Date())),
statement = c("income-statement-bulk",
"balance-sheet-statement-bulk",
"cash-flow-statement-bulk",
"ratios-bulk",
"key-metrics-bulk",
"financial-growth-bulk",
"income-statement-growth-bulk",
"balance-sheet-statement-growth-bulk",
"cash-flow-statement-growth-bulk"),
period = c("annual", "quarter")
) {
# DEBUG
# library(findata)
# library(httr)
# library(data.table)
# uri = "F:/equity/usa/fundamentals"
# years = 1990:(data.table::year(Sys.Date()))
# statement ="rating-bulk"
# period = "quarter"
# self <- FMP$new()
# define save_uri
save_uri = file.path(uri, statement, period)
if (!dir.exists(save_uri)) {
dir.create(save_uri, recursive = TRUE)
}
# prepare donwload data
url <- "https://financialmodelingprep.com/api/v4"
url <- paste(url, statement, sep = "/")
# get new years to download
years_exist = gsub(".csv", "", list.files(save_uri))
years = c(setdiff(years, years_exist), year(Sys.Date()))
# donwload data
lapply(years, function(year) {
print(year)
file_name = file.path(save_uri, paste0(year, ".csv"))
RETRY("GET",
url,
query = list(year = year, period = period, apikey = self$api_key),
write_disk(file_name, overwrite = TRUE))
return(NULL)
})
},
#' @description Help function for getting grades and targets
#'
#' @param symbols stock symbols.
#' @param tag url tag.
#'
#' @return Result of GET request
get_daily_v4 = function(symbols, tag = "price-target") {
# get data for tagfor all symbols
dt_l <- lapply(symbols, function(s) {
private$fmpv_path(
path = tag,
v = "v4",
symbol = s,
apikey = self$api_key
)
})
dt <- rbindlist(dt_l)
# clean data
dt <- na.omit(dt, c("symbol"))
dt <- unique(dt)
return(dt)
},
#' @description Get Upgrades and Downgrades from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#'
#' @return Result of GET request
get_grades = function(symbols) {
# get all grades for all symbols
grades = self$get_daily_v4(symbols, "upgrades-downgrades")
grades[, publishedDate := as.POSIXct(publishedDate,
format = "%Y-%m-%dT%H:%M:%OSZ",
tz = "UTC")]
return(grades)
},
#' @description Get Price Targest from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#'
#' @return Result of GET request
get_targets = function(symbols) {
# get all targets for all symbols
targets = self$get_daily_v4(symbols, "price-target")
targets[, publishedDate := as.POSIXct(publishedDate,
format = "%Y-%m-%dT%H:%M:%OSZ",
tz = "UTC")]
return(targets)
},
#' @description Get Ratings from FMP cloud.
#'
#' @param symbols stock symbols. If tiledburi, import symbols from FMP events.
#' @param uri uri to save output data.
#' @param update Should we scrape all data od jut the new one.
#' @param update_limit how many days to retrieve if update eqtual to TRUE.
#'
#' @return Result of GET request
get_ratings = function(symbols,
uri = "ratings.parquet",
update = FALSE,
update_limit = 10) {
# debug
# library(findata)
# library(httr)
# library(data.table)
# library(arrow)
# self = FMP$new()
# uri = "F:/equity/usa/fundamentals/ratings.parquet"
# private = list(
# ea_file_name = "EarningAnnouncements",
# transcripts_file_name = "earnings-calendar.rds",
# prices_file_name = "prices.csv",
# market_cap_file_name = "MarketCap",
#
# fmpv_path = function(path = "earning_calendar", v = "v3", ...) {
#
# # query params
# query_params <- list(...)
#
# # define url
# url <- paste0("https://financialmodelingprep.com/api/", v, "/", path)
#
# # get data
# p <- RETRY("GET", url, query = query_params, times = 5)
# result <- suppressWarnings(rbindlist(httr::content(p), fill = TRUE))
#
# return(result)
# }
# )
# get symbols for transcripts
# events = arrow::read_parquet("F:/equity/usa/fundamentals/earning_announcements.parquet")
# events = as.data.table(events)
# events <- events[date < Sys.Date()] # remove announcements for today
# events <- na.omit(events, cols = c("eps")) # remove rows with NA for earnings
# url <- modify_url("https://financialmodelingprep.com/", path = "api/v3/available-traded/list",
# query = list(apikey = Sys.getenv("APIKEY-FMPCLOUD")))
# res <- GET(url)
# stocks <- rbindlist(content(res), fill = TRUE)
# usa_symbols <- stocks[exchangeShortName %in% c("AMEX", "NASDAQ", "NYSE", "OTC")]
# events <- events[symbol %in% usa_symbols$symbo]
# symbols <- unique(events$symbol)
# help function
ratings_post = function(dt) {
ratings <- rbindlist(ratings_l)
ratings[, date := as.Date(date)]
# keep only numeric columns, without description
cols_ <- c("symbol", "date", "rating",
colnames(ratings)[grep("Score", colnames(ratings))])
ratings <- ratings[, ..cols_]
}
# get new data if update is FALSE
if (update == FALSE) {
# get all ratings for all symbols
ratings_l <- lapply(symbols, function(s) {
private$fmpv_path(
path = paste0("historical-rating/", s),
v = "v3",
limit = 10000,
apikey = self$api_key
)
})
ratings = ratings_post(ratings_l)
} else {
# import existing data
ratings_raw <- as.data.table(arrow::read_parquet(uri))
# get new data
ratings_l <- lapply(symbols, function(s) {
private$fmpv_path(
path = paste0("historical-rating/", s),
v = "v3",
limit = update_limit,
apikey = self$api_key
)
})
ratings = ratings_post(ratings_l)
# rbind
ratings = rbind(ratings_raw, ratings)
}
# clean data
ratings <- na.omit(ratings, c("symbol", "date"))
ratings <- unique(ratings)
# save to uri
arrow::write_parquet(ratings, uri)
return(NULL)
},
#' @description Get Historical dividends from FMP cloud.
#'
#' @param symbols Vector of stock symbols.
#'
#' @return Data table with historical dividends for symbol.
get_historical_dividends = function(symbols) {
# dividends
urls <- paste0("https://financialmodelingprep.com/api/v3/historical-price-full/stock_dividend/",
symbols)
dividends_l <- lapply(urls, function(x) {
p <- RETRY("GET", x, query = list(apikey = self$api_key))
res <- httr::content(p)
div <- rbindlist(res$historical)
cbind(symbol = res$symbol, div)
})
dividends <- rbindlist(dividends_l, fill = TRUE)
return(dividends)
},
#' @description Get hour data for all history from fmp cloud and save as parquet on AWS.
#'
#' @param symbols Symbol of the stock.
#' @param uri_minute AWS S3 bucket uri or NAS path.
#' @param save_uri_hour AWS S3 bucket uri for hour data.
#' @param save_uri_daily AWS S3 bucket uri for daily data.
#' @param deep_scan should we test for dates with low number od observation
#' and try to scrap again.
#'
#' @return Data saved to uri.
get_equities_batch = function(symbols,
uri_minute = "s3://equity-usa-minute",
deep_scan = FALSE) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(RcppQuantuccia)
# library(lubridate)
# library(nanotime)
# library(arrow)
# library(checkmate)
# library(future.apply)
# self = FMP$new()
# symbols = c("AAPL", "SPY")
# uri_minute = "F:/equity/usa/minute" # "s3://equity-usa-minute"
# deep_scan = FALSE
# fmp = FMP$new()
# library(rvest)
# sp100 <- read_html("https://en.wikipedia.org/wiki/S%26P_100") |>
# html_elements(x = _, "table") |>
# (`[[`)(3) |>
# html_table(x = _, fill = TRUE)
# sp100_symbols <- c("SPY", "TLT", "GLD", "USO", "SVXY",
# sp100$Symbol)
# symbols <- fmp$get_sp500_symbols()
# symbols <- na.omit(symbols)
# symbols <- symbols[!grepl("/", symbols)]
# symbols <- unique(union(sp100_symbols, symbols))
# check if we have all necessary env variables
assert_choice("AWS_ACCESS_KEY_ID", names(Sys.getenv()))
assert_choice("AWS_SECRET_ACCESS_KEY", names(Sys.getenv()))
assert_choice("AWS_DEFAULT_REGION", names(Sys.getenv()))
# # s3 bucket
# if (grepl("^s3:/", uri_minute)) {
# bucket = s3_bucket(uri_minute)
# dir_files = bucket$ls()
# } else {
# dir_files = list.files(uri_minute)
# }
# parallel execution
# plan("multisession", workers = 2L)
# ADD FUTURE LAPPLY BELOW
# main loop to scrap minute data
lapply(symbols, function(s) {
# debug
print(s)
# define file name and s3 file name
file_name = paste0(s, ".parquet")
file_name_full = file.path(uri_minute, file_name)
# define start_dates
start_dates <- seq.Date(as.Date("2004-01-01"), Sys.Date() - 1, by = 1)
start_dates <- start_dates[isBusinessDay(start_dates)]
# get trading days from daily data
print("Get daily data")
daily_start <- seq.Date(as.Date("2004-01-01"), Sys.Date(), by = 365 * 4)
daily_end <- c(daily_start[-1], Sys.Date())
daily_data <- lapply(seq_along(daily_start), function(i) {
daily_data <- tryCatch({
self$get_intraday_equities(s,
multiply = 1,
time = 'day',
from = daily_start[i],
to = daily_end[i])
}, error = function(e) NULL)
})
daily_data <- rbindlist(daily_data)
if (length(daily_data) == 0) return(NULL)
start_dates <- as.Date(intersect(start_dates, as.Date(daily_data$formated)), origin = "1970-01-01")
# TODO: check BRK-b (BTK.B) and other symbols with -/.
# OLD DATA ----------------------------------------------------------------
# read old data
if (file_name %in% dir_files) {
# read data for the symbol
print("Get minute data")
data_history = arrow::read_parquet(file_name_full)
# cont if there is history data
if (length(data_history) > 0) {
# basic clean
data_history <- unique(data_history, by = "date")
setorder(data_history, date)
# missing freq
if (deep_scan) { # CHECK THIS LATER !!!
# create date column
data_history_tz <- copy(data_history)
# data_history_tz[, time := as.POSIXct(time, origin = as.POSIXct("1970-01-01 00:00:00", tz = "UTC"), tz = "UTC")]
# data_history_tz[, time := with_tz(time, tz = "America/New_York")]
# data_history_tz[, time_ := as.Date(time, tz = "America/New_York")]
# data_history_tz[, time_n := .N, by = time_]
#
# # define dates we will try to scrap again
# observation_per_day <- 60 * 6
# try_again <- unique(data_history_tz[time_n < observation_per_day, time_])
# start_dates <- as.Date(intersect(try_again, start_dates), origin = "1970-01-01")
# end_dates <- start_dates
} else {
# define dates to scrap
data_history_date = data_history[, unique(as.Date(date, tz = "America/New_York"))]
# get final dates to scrap
start_dates <- as.Date(setdiff(start_dates, data_history_date), origin = "1970-01-01")
end_dates <- start_dates
# data_history[as.Date(date) == as.Date("2004-01-16")]
}
} else {
dates <- self$create_start_end_dates(start_dates, 1)
start_dates <- dates$start_dates
end_dates <- dates$end_dates
}
} else {
dates <- self$create_start_end_dates(start_dates, 1)
start_dates <- dates$start_dates
end_dates <- dates$end_dates
}
# GET NEW DATA ------------------------------------------------------------
# if there is no dates next
if (length(start_dates) == 0) {
print(paste0("No data for symbol ", s))
return(NULL)
}
# get data
print("Get new minute data")
system.time({
data_slice = lapply(seq_along(start_dates), function(d) {
# for (d in seq_along(start_dates[1:10])) {
if (end_dates[d] >= Sys.Date()) end_dates[d] <- Sys.Date() - 1
self$get_intraday_equities(
s,
multiply = 1,
time = "minute",
from = start_dates[d],
to = end_dates[d]
)
})
})
if (any(unlist(sapply(data_slice, nrow)) > 4999)) {
stop("More than 4999 rows!")
}
data_by_symbol <- rbindlist(data_slice)
# if there is no data next
if (nrow(data_by_symbol) == 0) {
print(paste0("No data for symbol ", s))
return(NULL)
}
# convert to numeric (not sure why I put these, but it have sense I believe).
cols <- c("o", "h", "l", "c", "v")
data_by_symbol[, (cols) := lapply(.SD, as.numeric), .SDcols = cols]
# clean data
data_by_symbol <- unique(data_by_symbol, by = c("formated"))
data_by_symbol[, date := as.POSIXct(t / 1000,
origin = as.POSIXct("1970-01-01 00:00:00", tz = "UTC"),
tz = "UTC")]
data_by_symbol = data_by_symbol[, .(date, o, h, l, c, v)]
setnames(data_by_symbol, c("date", "open", "high", "low", "close", "volume"))
# rbind new an old
if (file_name %in% dir_files && nrow(data_history) > 0) {
# rbind new data
data_by_symbol = rbind(data_history, data_by_symbol)
setorder(data_by_symbol, "date")
data_by_symbol = unique(data_by_symbol, by = "date")
# delete old parquet file
if (grepl("^s3:/", uri_minute)) {
bucket$DeleteFile(file_name)
} else {
file.remove(file_name_full)
}
# save file
arrow::write_parquet(data_by_symbol, file_name_full)
} else {
# save to S3
arrow::write_parquet(data_by_symbol, file_name_full)
}
return(NULL)
})
},
#' @description Get minute data for all history from FMP cloud.
#'
#' @param symbols Symbol of the stock.
#' @param uri_minute AWS S3 bucket uri or NAS path.
#' @param deep_scan should we test for dates with low number od observation
#' and try to scrap again.
#' @param workers Number of workers.
#'
#' @return Data saved to uri.
get_minute = function(symbols,
uri_minute,
deep_scan = FALSE,
workers = 1L) {
# debug
# library(findata)
# library(data.table)
# library(httr)
# library(RcppQuantuccia)
# library(lubridate)
# library(nanotime)
# library(arrow)
# library(checkmate)
# library(future.apply)
# self = FMP$new()
# symbols = c("AAPL", "SPY")
# uri_minute = "F:/equity/usa/minute"
# deep_scan = FALSE
# fmp = FMP$new()
# library(rvest)
# sp100 <- read_html("https://en.wikipedia.org/wiki/S%26P_100") |>
# html_elements(x = _, "table") |>
# (`[[`)(3) |>
# html_table(x = _, fill = TRUE)
# sp100_symbols <- c("SPY", "TLT", "GLD", "USO", "SVXY",
# sp100$Symbol)
# symbols <- fmp$get_sp500_symbols()
# symbols <- na.omit(symbols)
# symbols <- symbols[!grepl("/", symbols)]
# symbols <- unique(union(sp100_symbols, symbols))
# parallel execution
# plan("multisession", workers = 2L)
# ADD FUTURE LAPPLY BELOW
# main loop to scrap minute data
mclapply(symbols, function(s) {
# debug
# s = "WBA"
print(s)
# create folder for symbol if it doesn't exists
dir_name = file.path(uri_minute, s)
if (!dir.exists(dir_name)) {
dir.create(dir_name)
}
# get trading days from daily data
print("Get daily data")
daily_start <- seq.Date(as.Date("2004-01-01"), Sys.Date(), by = 365 * 4)
daily_end <- c(daily_start[-1], Sys.Date())
daily_data <- lapply(seq_along(daily_start), function(i) {
daily_data <- tryCatch({
self$get_intraday_equities(s,
multiply = 1,
time = 'day',
from = daily_start[i],
to = daily_end[i])
}, error = function(e) NULL)
})
daily_data <- rbindlist(daily_data)
if (length(daily_data) == 0) return(NULL)
dates <- as.Date(daily_data$formated)
# keep only NYSE business days
dates_business = getBusinessDays(min(dates), max(dates))[]
dates_business = dates_business[dates_business %in% dates]
dates = dates_business
# extract missing dates
dates_exists = gsub(".parquet", "", list.files(dir_name))
dates_new = as.Date(setdiff(paste0(dates), dates_exists))
# get new data
for (i in seq_along(dates_new)) {
# crate date
date = dates_new[i]
print(date)
# get new data for date
date_data = self$get_intraday_equities(
s,
multiply = 1,
time = "minute",
from = date,
to = date
)
# if there is no data next
if (nrow(date_data) == 0) {
print(paste0("No data for date ", date, " for symbol ", s))
date_data = data.table(
o = NA_real_,
h = NA_real_,
l = NA_real_,
c = NA_real_,
v = NA_real_,
t = NA_real_,
formated = "no minute data"
)
}
# save file
file_name_full = file.path(dir_name, paste0(date, ".parquet"))
arrow::write_parquet(date_data, file_name_full)
}
}, mc.cores = workers)
},
#' @description Get dividend data from FMP cloud Prep.
#'
#' @param ticker Ticker.
#'
#' @return data.table with dividend data.
get_dividends = function(ticker) {
url <- "https://financialmodelingprep.com/api/v3/historical-price-full/stock_dividend/"
url <- paste0(url, toupper(ticker))
p <- RETRY("GET", url, query = list(apikey = self$api_key))
res <- content(p)
res <- rbindlist(res$historical, fill = TRUE)
return(res)
},
#' @description Get beneficial ownership from FMP cloud.
#'
#' @param ticker Ticker.
#'
#' @return data.table with beneficial ownership data.
beneficial_ownership = function(ticker) {
# ticker = "AAPL"
# self$api_key <- Sys.getenv("APIKEY-FMPCLOUD")
url <- "https://financialmodelingprep.com/api/v4/insider/ownership/acquisition_of_beneficial_ownership"
p <- RETRY("GET", url, query = list(symbol = toupper(ticker), apikey = self$api_key))
res <- content(p)
res <- rbindlist(res, fill = TRUE)
return(res)
},
# QUOTE -------------------------------------------------------------------
#' @description Get the latest bid and ask prices for a stock, as well as the
#' volume and last trade price in real-time.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the full quote for the specified stock.
get_full_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/quote/", symbol)
p = RETRY("GET", url, query = list(apikey = self$api_key), times = 5)
res = content(p)
res = rbindlist(res, fill = TRUE)
res[, time := as.POSIXct(timestamp, origin = "1970-01-01", tz = "America/New_York")]
return(res)
},
#' @description Get a simplified quote for a stock, including the current price, volume, and last trade price.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the quote order for the specified stock.
get_quote_order = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/quote-order/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get a simple quote for a stock, including the price, change, and volume.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the simple quote for the specified stock.
get_simple_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0("https://financialmodelingprep.com/api/v3/quote-short/", symbol)
p = RETRY("GET", url, query = list(apikey = self$api_key), times = 5)
res = content(p)
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get the latest bid and ask prices for an OTC stock, as well as the volume and last trade price in real-time.
#'
#' @param symbol The OTC stock symbol(s) (e.g., "BATRB,FWONB").
#'
#' @return A data.table containing the OTC quote for the specified stock(s).
get_otc_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/otc/real-time-price/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get real-time prices for all stocks listed on a specific exchange.
#'
#' @param exchange The exchange for which to retrieve prices (e.g., "NYSE").
#'
#' @return A data.table containing real-time prices for all stocks on the specified exchange.
get_exchange_prices = function(exchange = "NYSE") {
assert_character(exchange, len = 1, null.ok = FALSE)
url = paste0("https://financialmodelingprep.com/api/v3/quotes/", exchange)
p = RETRY("GET", url, query = list(apikey = self$api_key), times = 5)
res = content(p)
res = rbindlist(res, fill = TRUE)
res[, time := as.POSIXct(timestamp, tz = "America/New_York")]
return(res)
},
#' @description Get the change in a stock's price over a given period of time.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing the stock price change for the specified stock.
get_stock_price_change = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v3/stock-price-change/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get information on trades that have occurred in the aftermarket.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing aftermarket trade information for the specified stock.
get_aftermarket_trades = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/pre-post-market-trade/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = as.data.table(res)
return(res)
},
#' @description Get the latest bid and ask prices for a stock in the aftermarket.
#'
#' @param symbol The stock symbol (e.g., "AAPL").
#'
#' @return A data.table containing aftermarket quote information for the specified stock.
get_aftermarket_quote = function(symbol) {
assert_character(symbol, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/pre-post-market/", symbol)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = as.data.table(res)
return(res)
},
#' @description Get quotes for multiple stocks at once.
#'
#' @param symbols A comma-separated string of stock symbols (e.g., "AAPL,MSFT").
#'
#' @return A data.table containing batch quote information for the specified stocks.
get_batch_quote = function(symbols) {
assert_character(symbols, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/batch-pre-post-market/", symbols)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
},
#' @description Get trades for multiple stocks at once.
#'
#' @param symbols A comma-separated string of stock symbols (e.g., "AAPL,MSFT").
#'
#' @return A data.table containing batch trade information for the specified stocks.
get_batch_trade = function(symbols) {
assert_character(symbols, len = 1, null.ok = FALSE)
url = paste0(self$base_url, "/v4/batch-pre-post-market-trade/", symbols)
res = content(RETRY("GET", url, query = list(apikey = self$api_key), times = 5))
res = rbindlist(res, fill = TRUE)
return(res)
}
),
private = list(
# PRIVATE -----------------------------------------------------------------
ea_file_name = "EarningAnnouncements",
transcripts_file_name = "earnings-calendar.rds",
prices_file_name = "prices.csv",
market_cap_file_name = "MarketCap",
fmpv_path = function(path = "earning_calendar", v = "v3", ...) {
# query params
query_params <- list(...)
# define url
url <- paste0("https://financialmodelingprep.com/api/", v, "/", path)
# get data
p <- RETRY("GET", url, query = query_params, times = 5)
result <- suppressWarnings(rbindlist(httr::content(p), fill = TRUE))
return(result)
}
)
)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.