Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice

Publikation: Working paperForskning

Standard

Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. / Callot, Laurent; Kock, Anders Bredahl; Medeiros, Marcelo C.

Aarhus : Institut for Økonomi, Aarhus Universitet, 2014.

Publikation: Working paperForskning

Harvard

Callot, L, Kock, AB & Medeiros, MC 2014 'Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice' Institut for Økonomi, Aarhus Universitet, Aarhus.

APA

Callot, L., Kock, A. B., & Medeiros, M. C. (2014). Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. Institut for Økonomi, Aarhus Universitet. CREATES Research Papers, Nr. 2014-42

CBE

Callot L, Kock AB, Medeiros MC. 2014. Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. Aarhus: Institut for Økonomi, Aarhus Universitet.

MLA

Callot, Laurent, Anders Bredahl Kock og Marcelo C. Medeiros Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. Aarhus: Institut for Økonomi, Aarhus Universitet. (CREATES Research Papers; Journal nr. 2014-42). 2014., 34 s.

Vancouver

Callot L, Kock AB, Medeiros MC. Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. Aarhus: Institut for Økonomi, Aarhus Universitet. 2014 nov 17.

Author

Callot, Laurent ; Kock, Anders Bredahl ; Medeiros, Marcelo C. / Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice. Aarhus : Institut for Økonomi, Aarhus Universitet, 2014. (CREATES Research Papers; Nr. 2014-42).

Bibtex

@techreport{2d29d52653444aea86d1eca138dc15ac,
title = "Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice",
abstract = "In this paper we consider modeling and forecasting of large realized covariance matrices by penalized vector autoregressive models. We propose using Lasso-type estimators to reduce the dimensionality to a manageable one and provide strong theoretical performance guarantees on the forecast capability of our procedure. To be precise, we show that we can forecast future 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 for the realized covariance matrices of the 30 Dow Jones stocks and find that these dynamics are not stable as the data is aggregated from the daily to the weekly and monthly frequency.The theoretical performance guarantees on our forecasts are illustrated on the Dow Jones index. In particular, we can beat our benchmark by a wide margin at the longer forecast horizons. 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 the forecasts.",
keywords = "Realized covariance, vector autoregression, shrinkage, Lasso, forecasting, portfolio allocation, Realized covariance, Vector autoregression, Shrinkage, Lasso, Forecasting, Portfolio allocation",
author = "Laurent Callot and Kock, {Anders Bredahl} and Medeiros, {Marcelo C.}",
year = "2014",
month = nov,
day = "17",
language = "English",
series = "CREATES Research Papers",
publisher = "Institut for {\O}konomi, Aarhus Universitet",
number = "2014-42",
type = "WorkingPaper",
institution = "Institut for {\O}konomi, Aarhus Universitet",

}

RIS

TY - UNPB

T1 - Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice

AU - Callot, Laurent

AU - Kock, Anders Bredahl

AU - Medeiros, Marcelo C.

PY - 2014/11/17

Y1 - 2014/11/17

N2 - In this paper we consider modeling and forecasting of large realized covariance matrices by penalized vector autoregressive models. We propose using Lasso-type estimators to reduce the dimensionality to a manageable one and provide strong theoretical performance guarantees on the forecast capability of our procedure. To be precise, we show that we can forecast future 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 for the realized covariance matrices of the 30 Dow Jones stocks and find that these dynamics are not stable as the data is aggregated from the daily to the weekly and monthly frequency.The theoretical performance guarantees on our forecasts are illustrated on the Dow Jones index. In particular, we can beat our benchmark by a wide margin at the longer forecast horizons. 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 the forecasts.

AB - In this paper we consider modeling and forecasting of large realized covariance matrices by penalized vector autoregressive models. We propose using Lasso-type estimators to reduce the dimensionality to a manageable one and provide strong theoretical performance guarantees on the forecast capability of our procedure. To be precise, we show that we can forecast future 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 for the realized covariance matrices of the 30 Dow Jones stocks and find that these dynamics are not stable as the data is aggregated from the daily to the weekly and monthly frequency.The theoretical performance guarantees on our forecasts are illustrated on the Dow Jones index. In particular, we can beat our benchmark by a wide margin at the longer forecast horizons. 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 the forecasts.

KW - Realized covariance, vector autoregression, shrinkage, Lasso, forecasting, portfolio allocation

KW - Realized covariance

KW - Vector autoregression

KW - Shrinkage

KW - Lasso

KW - Forecasting

KW - Portfolio allocation

M3 - Working paper

T3 - CREATES Research Papers

BT - Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice

PB - Institut for Økonomi, Aarhus Universitet

CY - Aarhus

ER -