midasml - Estimation and Prediction Methods for High-Dimensional Mixed Frequency Time Series Data

The 'midasml' package implements estimation and prediction methods for high-dimensional mixed-frequency (MIDAS) time-series and panel data regression models. The regularized MIDAS models are estimated using orthogonal (e.g. Legendre) polynomials and sparse-group LASSO (sg-LASSO) estimator. For more information on the 'midasml' approach see Babii, Ghysels, and Striaukas (2021, JBES forthcoming) <doi:10.1080/07350015.2021.1899933>. The package is equipped with the fast implementation of the sg-LASSO estimator by means of proximal block coordinate descent. High-dimensional mixed frequency time-series data can also be easily manipulated with functions provided in the package.

Last updated 2 years ago

forecasting-modelsmachine-learningnowcasting-modelssparse-group-lassotime-seriesfortran

4.27 score 37 stars 5 scripts 917 downloads

FAS - Factor-Augmented Sparse Regression Tuning-Free Testing

The 'FAS' package implements the bootstrap method for the tuning parameter selection and tuning-free inference on sparse regression coefficient vectors. Currently, the test could be applied to linear and factor-augmented sparse regressions, see Lederer & Vogt (2021, JMLR) <https://www.jmlr.org/papers/volume22/20-539/20-539.pdf> and Beyhum & Striaukas (2023) <arXiv:2307.13364>.

Last updated 1 years ago

fortran

2.70 score 1 scripts 160 downloads

LassoNet - 3CoSE Algorithm

Contains functions to estimate a penalized regression model using 3CoSE algorithm, see Weber, Striaukas, Schumacher Binder (2018) <doi:10.2139/ssrn.3211163>.

Last updated 5 years ago

cpp

1.70 score 138 downloads