Lassoed Boosting and Linear Prediction in Equities Market
Author:
Preprint
Abstract
We consider a two-stage estimation method for linear regression that uses the lasso in Tibshirani (1996) to screen variables and re-estimate the coefficients using the least-squares boosting method in Friedman (2001) on every set of selected variables. Based on the large-scale simulation experiment in Hastie et al. (2020), the performance of lassoed boosting is found to be as competitive as the relaxed lasso in Meinshausen (2007) and can yield a sparser model under certain scenarios. An application to predict equity returns also shows that lassoed boosting can give the smallest mean square prediction error among all methods under consideration.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- The Characteristics that Provide Independent Information about Average U.S. Monthly Stock ReturnsThe Review of Financial Studies, 2017
- SparseNet: Coordinate Descent With Nonconvex PenaltiesJournal of the American Statistical Association, 2011
- Relaxed LassoComputational Statistics & Data Analysis, 2007
- The Adaptive Lasso and Its Oracle PropertiesJournal of the American Statistical Association, 2006
- Boosting for high-dimensional linear modelsThe Annals of Statistics, 2006
- Model Selection and Estimation in Regression with Grouped VariablesJournal of the Royal Statistical Society Series B: Statistical Methodology, 2005
- Regularization and Variable Selection Via the Elastic NetJournal of the Royal Statistical Society Series B: Statistical Methodology, 2005
- Variable Selection via Nonconcave Penalized Likelihood and its Oracle PropertiesJournal of the American Statistical Association, 2001
- Smoothing Parameter Selection in Nonparametric Regression Using an Improved Akaike Information CriterionJournal of the Royal Statistical Society Series B: Statistical Methodology, 1998
- Regression Shrinkage and Selection Via the LassoJournal of the Royal Statistical Society Series B: Statistical Methodology, 1996