README.md

In johannestang/forecastexp: General purpose forecast experiment framework

forecastexp

An R package for conducting macroeconomic forecasting experiments.

Content

This package contains a collection of code needed to replicate the material in some of my papers. Note: Despite being collected in a package the code has been written specifically for these papers and contain very little error checking, so if you choose to use the code for other purposes do so with caution.

Usage

The package requires the macrods package, both are easily installed using devtools:

library(devtools)
install_github('johannestang/macrods')
install_github('johannestang/forecastexp')
library(forecastexp)

In addition the following packages should be installed : glmnet, lars, pryr, quantreg, xtable and zoo, all available from CRAN.

The function replicate_swjbes gives an example of how to use the package. It replicates part of the results in Stock, J. and M. Watson (2002): "Forecasting using principal components from a large number of predictors," Journal of the American Statistical Association, 97, 1167–1179.

For further examples see sparsedi-rep and l1factor-rep.

johannestang/forecastexp documentation built on May 19, 2019, 3:02 p.m.