In msperlin/adfeR: Data, exercises and Functions for Book "Analise de Dados Financeiros e Economicos com o R"

library(BatchGetSymbols)
library(purrr)

ticker <- '^GSPC'
first_date <- '1950-01-01'
last_date <- '2021-01-01'
df_SP500 <- BatchGetSymbols(tickers = '^GSPC', 
                            first.date = first_date,
                            last.date = last_date)[[2]]

replace_outliers <- function(col_in, my_prob = 0.05) {
  # Replaces outliers from a vector
  #
  # INPUTS: col_in The vector
  #         my_prob Probability of quantiles (p and 1-p)
  #
  # OUTPUT: A vector

  # return if class is other than numeric
  if (!(class(col_in) %in% c('numeric', 'integer'))) return(col_in)

  my_outliers <- stats::quantile(x = col_in,
                                 probs = c(my_prob, 1-my_prob),
                                 na.rm = TRUE)

  idx <- (col_in <= my_outliers[1])|(col_in >= my_outliers[2])
  col_in[idx] <- NA

  return(col_in)

}

# remove outlivers from vectors
l_out <- map(df_SP500, replace_outliers, my_prob = 0.025)

df_SP500_nooutlier <- na.omit(as_tibble(l_out))

nrow_1 <- nrow(df_SP500)
nrow_2 <- nrow(df_SP500_nooutlier)

my_sol <- nrow_1 - nrow_2

# none
my_answers <- make_random_answers(my_sol)

Question

Use função replace_outliers (veja seção \@ref(fct-replace)), criada neste capítulo, para remover outliers de todas as colunas numéricas dos dados do SP500 importados anteriormente com my_prob = 0.025. Quantas linhas foram perdidas neste processo de limpeza?

exams::answerlist(my_answers, markup = "markdown")

Solution

Meta-information

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

msperlin/adfeR documentation built on March 26, 2021, 3:05 a.m.