"We consider various MIDAS (Mixed Data Sampling) regression models to predict volatility. The models differ in the specification of regressors (squared returns, absolute returns, realized volatility, realized power, and return ranges), in the use of daily or intra-daily (5-minute) data, and in...

A typical MIDAS regression involves estimating parameters via nonlinear least squares, unless U-MIDAS is applied - which involves OLS - the latter being appealing when the sampling frequency differences are small. In this paper we propose to use OLS estimation of the MIDAS regression slope and...

This paper introduces structured machine learning regressions for high-dimensional time series data potentially sampled at different frequencies. The sparse-group LASSO estimator can take advantage of such time series data structures and outperforms the unstructured LASSO. We establish oracle...