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.
-
Install Julia.
-
Start Julia and run the following command at the
julia>
prompt:
using Pkg; Pkg.add(url="https://github.com/jwmi/BilinearModels.jl")
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)
A tutorial is provided in the Jupyter notebook here. Or, for a simple HTML rendering of the tutorial, see here.
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).
This package is released under the MIT "Expat" license. See LICENSE.md.