R/process_one_stock_ticker.R

In misrori/aranykez: Everything related to stock and crypto trading

#' Downloading and adding technical indicators to the historical data of a ticker
#' @export
#' @param ticker The ticker of the company to download
#' @param start_date The first date of the historical data
#' @param end_date The last date of the data
#' @param mas List of the simple moving averages to calculate
#' @param ad_local_text if you want to calculate the local min max with 20 period
#' @param ad_percent_change if you want to calculate the percent change from the first day
#' @importFrom  data.table data.table
#' @importFrom rtsdata ds.getSymbol.yahoo
#' @importFrom TTR RSI
#' @importFrom pracma movavg
stock_get_one_ticker  <- function(ticker, start_date = "2000-01-01", end_date = Sys.Date(),  mas=c(50, 100, 200), ad_local_text=F, ad_percent_change=F) {
  tryCatch({
    df <- data.frame(ds.getSymbol.yahoo(ticker, from = (as.Date(start_date)-250), to =end_date ))
    names(df) <- tolower(sapply(strsplit(names(df), '.', fixed = T), '[[', 2))
    df$date <- as.Date(row.names(df))
    row.names(df) <- 1:nrow(df)
    df <- data.table(df)

    if( !identical(names(df) , c("open","high","low", "close","volume",  "adjusted","date"))) {
      text<- paste0('Error: ', ticker, ' # problem: names of dataframe is bad ', ' time: ', Sys.time())
      stop(text)
    }

  }, error=function(x) {
    print(x)
    stop('No ticker')
  })

  for (simple_mas in mas) {
    if (nrow(df)<=simple_mas) {
      df[[paste0('ma_', simple_mas, '_value')]] <- 0

    }else{
      df[[paste0('ma_', simple_mas, '_value')]] <- movavg( df[['close']], simple_mas,type = 's' )
      df[[paste0('diff_',simple_mas,'_ma_value')]] <-  (( df[["close"]]  /df[[paste0('ma_', simple_mas, '_value')]] )-1)*100
    }
  }

  df$rsi <- RSI(as.numeric(df$high),n = 14)
  df <- df[date >= as.Date(start_date) & !is.na(rsi),]
  if (ad_local_text) {
    df <- utility_ad_local_min_max(df, 20)
  }
  if (ad_percent_change) {
    first_day_close <- df[!is.na(close) & close>0]$close[1]
    df$percent_change <- round(((df$close /first_day_close) -1 )*100,2)
  }
  df$ticker <- ticker

  return(df)
}