Versatile HAR model for realized volatility: A least square model averaging perspective
A rapidly growing body of literature has documented improvements in forecasting financial return volatility measurement using various heterogeneous autoregression (HAR) type models. Most HAR-type models use a fixed lag index of (1,5,22) to mirror the daily, weekly, and monthly components of the vola...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
KeAi Communications Co., Ltd.
2019-03-01
|
Series: | Journal of Management Science and Engineering |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2096232019300046 |
id |
doaj-d096e706c76f49a18321d6742938e04b |
---|---|
record_format |
Article |
spelling |
doaj-d096e706c76f49a18321d6742938e04b2020-11-25T03:48:45ZengKeAi Communications Co., Ltd.Journal of Management Science and Engineering2096-23202019-03-01415573Versatile HAR model for realized volatility: A least square model averaging perspectiveYue Qiu0Xinyu Zhang1Tian Xie2Shangwei Zhao3WISE and School of Economics, Xiamen University, Xiamen, Fujian, 361005, ChinaAcademy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, 100190, China; Department of Statistics, Pennsylvania State University, University Park, Pennsylvania 16802, USA; Corresponding author. Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, 100190, China.WISE and School of Economics, Xiamen University, Xiamen, Fujian, 361005, ChinaCollege of Science, Minzu University of China, Beijing, 100081, ChinaA rapidly growing body of literature has documented improvements in forecasting financial return volatility measurement using various heterogeneous autoregression (HAR) type models. Most HAR-type models use a fixed lag index of (1,5,22) to mirror the daily, weekly, and monthly components of the volatility process, but they ignore model specification uncertainty. In this paper, we propose applying the least squares model averaging approach to HAR-type models with signed realized semivariance to account for model uncertainty and to allow for a more flexible lag structure. We denote this approach as MARS and prove that the MARS estimator is asymptotically optimal in the sense of achieving the lowest possible mean squared forecast error. Selected by the data-driven model averaging method, the lag combination in the MARS method changes with various data series and different forecast horizons. Employing high frequency data from the NASDAQ 100 index and its 104 constituents, our empirical results demonstrate that acknowledging model uncertainty under the HAR framework and solving with the model averaging method can significantly improve the accuracy of financial return volatility forecasting. Keywords: Heterogeneous autoregression, Volatility forecasting, Forecasting combination, Model averaging, Asymptotic optimalityhttp://www.sciencedirect.com/science/article/pii/S2096232019300046 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Yue Qiu Xinyu Zhang Tian Xie Shangwei Zhao |
spellingShingle |
Yue Qiu Xinyu Zhang Tian Xie Shangwei Zhao Versatile HAR model for realized volatility: A least square model averaging perspective Journal of Management Science and Engineering |
author_facet |
Yue Qiu Xinyu Zhang Tian Xie Shangwei Zhao |
author_sort |
Yue Qiu |
title |
Versatile HAR model for realized volatility: A least square model averaging perspective |
title_short |
Versatile HAR model for realized volatility: A least square model averaging perspective |
title_full |
Versatile HAR model for realized volatility: A least square model averaging perspective |
title_fullStr |
Versatile HAR model for realized volatility: A least square model averaging perspective |
title_full_unstemmed |
Versatile HAR model for realized volatility: A least square model averaging perspective |
title_sort |
versatile har model for realized volatility: a least square model averaging perspective |
publisher |
KeAi Communications Co., Ltd. |
series |
Journal of Management Science and Engineering |
issn |
2096-2320 |
publishDate |
2019-03-01 |
description |
A rapidly growing body of literature has documented improvements in forecasting financial return volatility measurement using various heterogeneous autoregression (HAR) type models. Most HAR-type models use a fixed lag index of (1,5,22) to mirror the daily, weekly, and monthly components of the volatility process, but they ignore model specification uncertainty. In this paper, we propose applying the least squares model averaging approach to HAR-type models with signed realized semivariance to account for model uncertainty and to allow for a more flexible lag structure. We denote this approach as MARS and prove that the MARS estimator is asymptotically optimal in the sense of achieving the lowest possible mean squared forecast error. Selected by the data-driven model averaging method, the lag combination in the MARS method changes with various data series and different forecast horizons. Employing high frequency data from the NASDAQ 100 index and its 104 constituents, our empirical results demonstrate that acknowledging model uncertainty under the HAR framework and solving with the model averaging method can significantly improve the accuracy of financial return volatility forecasting. Keywords: Heterogeneous autoregression, Volatility forecasting, Forecasting combination, Model averaging, Asymptotic optimality |
url |
http://www.sciencedirect.com/science/article/pii/S2096232019300046 |
work_keys_str_mv |
AT yueqiu versatileharmodelforrealizedvolatilityaleastsquaremodelaveragingperspective AT xinyuzhang versatileharmodelforrealizedvolatilityaleastsquaremodelaveragingperspective AT tianxie versatileharmodelforrealizedvolatilityaleastsquaremodelaveragingperspective AT shangweizhao versatileharmodelforrealizedvolatilityaleastsquaremodelaveragingperspective |
_version_ |
1724497270267707392 |