Giter Site home page Giter Site logo

minhdg00 / penalized-mle-with-reparameterization Goto Github PK

View Code? Open in Web Editor NEW
0.0 2.0 0.0 480 KB

implement BCD to find l1-penalized MLE with appropriate parametrization multivariate GLM

R 74.10% C 25.90%
coordinate-descent-algorithm estimation algorithm lasso variable-selection reparametrized-gaussian-models

penalized-mle-with-reparameterization's Introduction

Research project supervised by Prof. Sunyoung Shin.

• Proposed a penalized Gaussian maximum likelihood estimation and variable selection method with appropriate reparameterization in multiple response regression model, and wrote code for this algorithm

Main idea of the algorithm:

Make effort to provide a efficient Lasso type estimator with parameterization tricks. The algorithm's purpose is to tackle with sparsity and large number of features.

File Instruction

  • dmlsim.R  is the simulation script
  • JointSol_rep.R is revised DML algorithm by Lee & Liu to fit our case
  • UpdatePhi_rep.c is the script to update the new parameter Phi
  • cv_dml.R is the script containing cross validation as well as dmlpath function ( fit model for a vector of lambdas)

Note

  • To run R code, you first need to compile the C code on your local machine
  • With large size of data, the code can run very slow.

Reference

Städler, Nicolas & Bühlmann, Peter & van de Geer, Sara. (2010). L1-Penalization for Mixture Regression Models. TEST: An Official Journal of the Spanish Society of Statistics and Operations Research. 19. 209-256. 10.1007/s11749-010-0197-z.
Lee, Wonyul & Liu, Yufeng. (2012). Simultaneous Multiple Response Regression and Inverse Covariance Matrix Estimation via Penalized Gaussian Maximum Likelihood. Journal of multivariate analysis. 111. 241-255. 10.1016/j.jmva.2012.03.013. 

penalized-mle-with-reparameterization's People

Contributors

minhdg00 avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.