Modeling and Forecasting Large Realized Covariance Matrices and Portfolio Choice

By Laurent Callot, Anders Bredahl Kock, Marcelo C. Medeiros. 2017

94 citations recorded by the source corpus.

Summary We consider modeling and forecasting large realized covariance matrices by penalized vector autoregressive models. We consider Lasso-type estimators to reduce the dimensionality and provide strong theoretical guarantees on the forecast capability of our procedure. We show that we can forecast realized covariance matrices almost as precisely as if we had known the true driving dynamics of these in advance. We next investigate the sources of these driving dynamics as well as the performance of the proposed models for forecasting the realized covariance matrices of the 30 Dow Jones stocks. We find that the dynamics are not stable as the data are aggregated from the daily to lower frequencies. Furthermore, we are able beat our benchmark by a wide margin. Finally, we investigate the economic value of our forecasts in a portfolio selection exercise and find that in certain cases an investor is willing to pay a considerable amount in order get access to our forecasts. Copyright © 2016 John Wiley & Sons, Ltd.

Subfields: Finance, Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/e00aa25a59fc60e489f7d16b6290eb99296cfc1b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Finance

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Forecasting realized volatility of agricultural commodity futures with infinite Hidden Markov HAR models (2019, unknown): “Based on Fleming & Hern´andez-Hern´andez (2003), Bollerslev et al. (2016), and Callot et al. (2017), we use the quadratic utility with risk aversion to determine the economic values of various forecast models: In this paper, we consider two levels of risk aversion rates for the investor, the mild…”
A Nodewise Regression Approach to Estimating Large Portfolios (2016, methodology): “(2018), and then eigenvalue cleaning as in Callot et al. (2017), and Hautsch et al. (2012). Note that we do not use this procedure anywhere in our simulations or applications. 1. Use the matrix symmetrization procedure of Fan et al. (2018): let Θ̂jk represent jth row and k th column of p× p matrix.”
Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (2020, background): “It has many important applications [54] such as forecasting the demand for items sold by retailers [41, 190, 156, 136, 14, 25], the flow of traffic [111, 126, 118], the demand and supply of energy [45, 170, 117, 157], or the covariance matrix, volatility and long-tail distributions in finance [30, 2”