tests/testthat/test-tq_transmute.R

library(tidyquant)
library(tidyverse)
context("Testing tq_transmute")

#### Setup ----
AAPL <- tq_get("AAPL", get = "stock.prices", from = "2010-01-01", to = "2015-01-01")

# Test 1: tq_transmute to.period
test1 <- AAPL %>%
    tq_transmute(select = close, mutate_fun = to.period, period = "months")

# Test 1.2: Grouped_df test
grouped_df <- grouped_df <- tibble(symbol = c("META", "AMZN")) %>%
    tq_get(
        get  = "stock.prices",
        from = "2015-01-01",
        to   = "2016-01-01"
    ) %>%
    group_by(symbol)

test1.2a  <- mutate(grouped_df, V1 = runSD(adjusted)) %>%
    select(-(open:adjusted))

test1.2b <- tq_transmute(grouped_df, adjusted, runSD, col_rename = "V1")


# Test 2: tq_transmute_xy test
test2 <- AAPL %>%
    tq_transmute_xy(x = close, mutate_fun = to.period, period = "months")

test2b <- grouped_df %>%
    tq_transmute_xy(x = close, y = open, mutate_fun = runCor, n = 5)


# Test 3: Test transmute hourly data / Test transmute timezone data
# time_index <- seq(from = as.POSIXct("2012-05-15 07:00"),
#                   to   = as.POSIXct("2012-05-17 18:00"),
#                   by   = "hour")
# set.seed(1)
# value <- rnorm(n = length(time_index))
# tz <- "Zulu"
# test3 <- tibble(time_index, value) %>%
#     dplyr::mutate(time_index = lubridate::ymd_hms(time_index, tz = tz)) %>%
#     tq_transmute_xy(x = value, mutate_fun = MACD)

# Test 4: transmute to.monthly which returns character dates
test4 <- AAPL %>%
    tq_transmute(select = NULL, mutate_fun = to.monthly)

# Test 5: Check tq_transmute_data
fb_returns <- tq_get("META", get  = "stock.prices", from = "2016-01-01", to   = "2016-12-31") %>%
    tq_transmute(adjusted, periodReturn, period = "monthly", col_rename = "fb.returns")
xlk_returns <- tq_get("XLK", from = "2016-01-01", to = "2016-12-31") %>%
    tq_transmute(adjusted, periodReturn, period = "monthly", col_rename = "xlk.returns")
test5 <- left_join(fb_returns, xlk_returns, by = "date")
regr_fun <- function(data) {
    coef(lm(fb.returns ~ xlk.returns, data = as.data.frame(data)))
}
test5 <- test5 %>%
    tq_transmute(mutate_fun = rollapply,
                 width      = 6,
                 FUN        = regr_fun,
                 by.column  = FALSE,
                 col_rename = c("coef.0", "coef.1"))

#### Tests ----

test_that("Test 1 returns tibble with correct rows and columns.", {
    # Tibble
    expect_is(test1, "tbl")
    # Rows
    expect_equal(nrow(test1), 60)
    # Columns
    expect_equal(ncol(test1), 2)
})

test_that("Test 1.2 grouped data frames are same with mutate and tq_transmute", {
    # Return column
    expect_identical(test1.2a$V1, test1.2b$V1)
    # Groups
    expect_identical(dplyr::groups(test1.2a), dplyr::groups(test1.2b))
})

test_that("Test 2 returns tibble with correct rows and columns.", {
    # Tibble
    expect_is(test2, "tbl")
    # Rows
    expect_equal(nrow(test2), 60)
    # Columns
    expect_equal(ncol(test2), 2)
})

# test_that("Test 3 returns tibble with correct rows and columns.", {
#     # Tibble
#     expect_is(test3, "tbl")
#     # Rows
#     expect_equal(nrow(test3), 60)
#     # Columns
#     expect_equal(ncol(test3), 3)
# })

# test_that("Test 3 returns correct timezone.", {
#     expect_equal({test3$time_index %>% lubridate::tz()}, tz)
# })

test_that("Test 5 returns tibble with correct rows and columns.", {
    # Tibble
    expect_is(test5, "tbl")
    # Rows
    expect_equal(nrow(test5), 12)
    # Columns
    expect_equal(ncol(test5), 3)
    # Check colnames
    expect_equal(colnames(test5), c("date", "coef.0", "coef.1"))
})

# Invalid data inputs
test_that("Test error on invalid data inputs.", {

    # Non-data.frame objects
    a <- seq(1:100)
    expect_error(
        seq(1:100) %>%
            tq_transmute(select = NULL, mutate_fun = to.monthly)
    )
    expect_error(
        seq(1:100) %>%
            tq_mutate_xy(x = a, mutate_fun = to.monthly)
    )

    # No date columns
    expect_error(
        tibble(a = seq(1:100)) %>%
            tq_transmute(select = NULL, mutate_fun = to.monthly),
        "No date or POSIXct column found in `data`."
    )
    expect_error(
        tibble(a = seq(1:100)) %>%
            tq_mutate_xy(x = a, mutate_fun = to.monthly),
        "No date or POSIXct column found in `data`."
    )
})

# Invalid select, x and y inputs
test_that("Test error on invalid select, x and y inputs.", {

    expect_error(
        {select <- "err"
        AAPL %>%
            tq_mutate_(select = select, mutate_fun = "to.monthly")}
    )
    expect_error(
        {x <-  "err"
        AAPL %>%
            tq_mutate_xy_(x = x, y = "close", mutate_fun = "Delt", k = 1)},
        paste0("x = err not a valid name.")
    )
    expect_error(
        {y <-  "err"
        AAPL %>%
            tq_mutate_xy_(x = "open", y = y, mutate_fun = "Delt", k = 1)},
        paste0("y = err not a valid name.")
    )

})

# Invalid mutate_fun, x and y inputs
test_that("Test error on invalid mutate_fun, x and y inputs.", {

    expect_error(
        {mutate_fun <- "err"
        AAPL %>%
            tq_mutate_(select = "close", mutate_fun = mutate_fun)},
        paste0("fun = err not a valid option.")
    )

})

# Invalid col_rename, can't have duplicate names
test_that("Test error on invalid col_rename, duplicate names.", {

    expect_error({
        col_rename <- c("name1", "name2", "name1")
        AAPL %>%
            tq_mutate_(select = "close", mutate_fun = "lag.xts",
                       k = 1:3, col_rename = col_rename)
        },
        paste0("Could not rename columns. Do you have duplicate names in `col_rename`?")
    )

})

Try the tidyquant package in your browser

Any scripts or data that you put into this service are public.

tidyquant documentation built on April 3, 2023, 5:13 p.m.