High dimensional generalized linear models for temporal dependent data
Authors:
- 1 February 2023
- journal article
- research article
- Published by Bernoulli Society for Mathematical Statistics and Probability in Bernoulli
- Vol. 29 (1) , 105-131
Abstract
High dimensional generalized linear models are widely applicable in many scientific fields with data having heavy tails. However, little is known about statistical guarantees on the estimates of such models in a time series setting. In this article, we establish statistical error bounds and support recovery guarantees of the classical ℓ1 regularized procedure for generalized linear model with temporal dependent data. We also propose a new robust M-estimator for high dimensional time series. Properties of the proposed robust procedure are investigated both theoretically and numerically. As an extension, we introduce a robust estimator for linear regression and show that the proposed robust estimator achieves nearly the optimal rate as that for i.i.d sub-Gaussian data. Simulation results show that the proposed method performs well numerically in the presence of heavy-tailed and serially dependent covariates and/or errors, and it significantly outperforms the classical Lasso method. For applications, we demonstrate, in the supplementary material, the regularized robust procedure via analyzing high-frequency trading data in finance.Keywords
This publication has 67 references indexed in Scilit:
- Oracle inequalities for the lasso in the Cox modelThe Annals of Statistics, 2013
- A Unified Framework for High-Dimensional Analysis of $M$-Estimators with Decomposable RegularizersStatistical Science, 2012
- Quasi-Likelihood and/or Robust Estimation in High DimensionsStatistical Science, 2012
- A Bernstein type inequality and moderate deviations for weakly dependent sequencesProbability Theory and Related Fields, 2010
- Spatio-temporal correlations and visual signalling in a complete neuronal populationNature, 2008
- Asymptotic spectral theory for nonlinear time seriesThe Annals of Statistics, 2007
- Multiple neural spike train data analysis: state-of-the-art and future challengesNature Neuroscience, 2004
- The Generalized LASSOIEEE Transactions on Neural Networks, 2004
- Short-term load forecasting via ARMA model identification including non-gaussian process considerationsIEEE Transactions on Power Systems, 2003
- The Influence Curve and its Role in Robust EstimationJournal of the American Statistical Association, 1974