Financial Forecasting Tool

Marc Boulet
2017-06-25

Purpose

  • Would you like to understand how you can try to beat the financial markets?
  • This app will let you test the forecast package in R on both historical and future data

Data source

Monthly close data retrieved from https://finance.yahoo.com/.

  • Dow Jones Industrial Average (February 1985 - June 2017)
  • Standard & Poors 500 (February 1950 - June 2017)
  • Toronto Stock Exchange (July 1979 - June 2017)

plot of chunk unnamed-chunk-1plot of chunk unnamed-chunk-1plot of chunk unnamed-chunk-1

Forecasting package

  • default option uses exponential smoothing method
  • forecasts produced using exponential smoothing methods are weighted averages of past observations, with the weights decaying exponentially as the observations get older.
  • For more information, please see: https://www.otexts.org/fpp.

Forecasting example

library(forecast)
plot (forecast(dji.ts, h=24), main="Dow Jones Industrial Average")

plot of chunk unnamed-chunk-2

  • Three parameter adjustments: input data, number of months to forecast and confidence intervals