In msperlin/afedR: Data and Functions for Book "Analyzing Financial and Economical Data with R"

library(BatchGetSymbols)
library(tidyverse)

tickers <- GetSP500Stocks()$Tickers
first_date <- Sys.Date() - 3*365
last_date  <- Sys.Date()

df_stocks <- BatchGetSymbols(tickers = tickers, 
                             first.date = first_date, 
                             last.date = last_date)[[2]]

df_sp500 <- BatchGetSymbols(tickers =  '^GSPC', 
                            first.date = first_date, 
                            last.date = last_date)[[2]]

library(dplyr)

my_tab <- df_stocks %>%
  group_by(ticker) %>%
  do(my_arima = arima(x = .$ret.adjusted.prices, 
                      order = c(1,0,0))) %>%
  mutate(arima_forecast = predict(my_arima, n.ahead = 1 )$pred[1])

glimpse(my_tab)

# solution
idx <- which.max(my_tab$arima_forecast )
print(my_tab$ticker[idx])

# none
#my_answers <- make_random_answers(my_sol)
my_answers <- rep(NA, 5)

Question

Using the tidyverse functions, dplyr::group_by and dplyr::do, estimate an ARIMA model for the returns of each stock, available from the import process of previous exercise. In the same output dataframe, create a new column with the forecast in t + 1 for each model. Which stock has the highest expected return for t + 1?

Solution

Meta-information

extype: string exsolution: r mchoice2string(c(TRUE, FALSE, FALSE, FALSE, FALSE), single = TRUE) exname: "function 01" exshuffle: TRUE

msperlin/afedR documentation built on Sept. 11, 2022, 9:49 a.m.