Variable selection in partially time-varying coefficient models
Authors:
- 1 July 2009
- journal article
- research article
- Published by Taylor & Francis in Journal of Nonparametric Statistics
- Vol. 21 (5) , 553-566
Abstract
A partially time-varying coefficient model is introduced to characterise the nonlinearity and trending phenomenon. To enhance predictability and to select significant variables in the parametric component of the model, the penalised least squares method with the help of the profile local linear technique is developed in this article. The convergence rate and the oracle property of the resulting estimator are established under mild conditions. A remarkable achievement of our results is that it does not require undersmoothing of the nonparametric component. Meanwhile, some extensions of the proposed model and method are also discussed. Furthermore, some numerical examples are provided to show that our theory and method work well in practice.Keywords
This publication has 32 references indexed in Scilit:
- Profile-kernel likelihood inference with diverging number of parametersThe Annals of Statistics, 2008
- One-step sparse estimates in nonconcave penalized likelihood modelsThe Annals of Statistics, 2008
- Variable selection in semiparametric regression modelingThe Annals of Statistics, 2008
- Statistical challenges with high dimensionality: feature selection in knowledge discovery2007
- Trending time-varying coefficient time series models with serially correlated errorsJournal of Econometrics, 2007
- Profile likelihood inferences on semiparametric varying-coefficient partially linear modelsBernoulli, 2005
- Semiparametric Non-Linear Time Series Model SelectionJournal of the Royal Statistical Society Series B: Statistical Methodology, 2004
- Asset Pricing with Conditioning Information: A New TestThe Journal of Finance, 2003
- Trending time series and macroeconomic activity: Some present and future challengesJournal of Econometrics, 2001
- Quasi-likelihood Estimation in Semiparametric ModelsJournal of the American Statistical Association, 1994