Fast Estimation of Multinomial Logit Models: R Package mnlogit

2014/04/11 by Asad Hasan, Zhiyu Wang, Hasan, Asad +3 · 1 citation
Mathematics · Computer Science · #Statistical Methods and Inference #Statistical Methods and Bayesian Inference #Bayesian Methods and Mixture Models

paper · pdf · doi:10.48550/arxiv.1404.3177

Abstract

We present R package mnlogit for training multinomial logistic regression models, particularly those involving a large number of classes and features. Compared to existing software, mnlogit offers speedups of 10x-50x for modestly sized problems and more than 100x for larger problems. Running mnlogit in parallel mode on a multicore machine gives an additional 2x-4x speedup on up to 8 processor cores. Computational efficiency is achieved by drastically speeding up calculation of the log-likelihood function's Hessian matrix by exploiting structure in matrices that arise in intermediate calculations.

Citations

• On the limited memory BFGS method for large scale optimization

Cited by

• Stochastic gradient descent methods for estimation with large data sets

Related

• logitr: Fast Estimation of Multinomial and Mixed Logit Models with Preference Space and Willingness to Pay Space Utility Parameterizations
• PIANO: A Fast Parallel Iterative Algorithm for Multinomial and Sparse Multinomial Logistic Regression
• DS-MLR: Exploiting Double Separability for Scaling up Distributed Multinomial Logistic Regression
• Multinomial Logistic Regression Algorithms via Quadratic Gradient
• The Information Theoretically Efficient Model (ITEM): A model for computerized analysis of large datasets