NeuralNetworkEconomics.jl¶

This package focuses on the use of neural networks (and other machine learning methods) in economics.

The notes on machine learning in economics (1, 2, 3, 4) were originally written for ECON 628. They remain a good overview and valuable list of references, but the code is all in R. If you want to use some existing methods based on the research of Chernozhukov and coauthors or Athey and coauthors, then it makes a lot of sense to use the R packages they developed (hdm and grf respectively). However, if you want to write code to do something new, it likely makes more sense to use Julia.

A brief review of Julia packages for machine learning (with examples focused on lasso) is in ml-julia.

The notes on neural networks (1, [2], … ) feature examples in Julia using Flux.jl.

<!–` –>