BilinearModels

About

BilinearModels is a Julia package for estimation and inference in generalized bilinear models.

Please cite the following publication if you use this package in your research:

J. W. Miller and S. L. Carter, Inference in generalized bilinear models. arXiv preprint arXiv:2010.04896, 2020.

Installation

Install Julia.
Start Julia and run the following command at the julia> prompt:

using Pkg; Pkg.add(url="https://github.com/jwmi/BilinearModels.jl")

Quick start

At the julia> prompt:

using BilinearModels

Y = [1 2 3; 4 5 6; 7 8 0]  # toy data matrix
I,J = size(Y)              # dimensions of Y
X = ones(I,1)              # feature covariate matrix
Z = ones(J,1)              # sample covariate matrix
M = 0                      # number of latent factors

# fit the model
A,B,C,D,U,V,S,T,omega,logp = BilinearModels.fit(Y,X,Z,M)  

# compute standard errors
se_A,se_B,se_C,se_U,se_V,se_S,se_T = BilinearModels.infer(Y,X,Z,A,B,C,D,U,V,S,T,omega)

Tutorial

A tutorial is provided in the Jupyter notebook here. Or, for a simple HTML rendering of the tutorial, see here.

Questions and issues

Please feel free to post any issues here, or submit a pull request if you make improvements to the code. For general questions, feel free to contact me (Jeff Miller).

Licensing

This package is released under the MIT "Expat" license. See LICENSE.md.