國立虎尾科技大學 |

Three Essays on Realized Volatility Models for High-Frequency Data.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Three Essays on Realized Volatility Models for High-Frequency Data./
Author:	Shen, Ji.
Description:	1 online resource (118 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 78-10(E), Section: A.
Contained By:	Dissertation Abstracts International78-10A(E).
Subject:	Economics. -
Online resource:	click for full text (PQDT)
ISBN:	9781369857016

Three Essays on Realized Volatility Models for High-Frequency Data.
Shen, Ji.

Three Essays on Realized Volatility Models for High-Frequency Data. - 1 online resource (118 pages)

Source: Dissertation Abstracts International, Volume: 78-10(E), Section: A.

Thesis (Ph.D.)

Includes bibliographical references

The dissertation centers on modeling volatility of stock returns with high frequency data. It contains three chapters.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9781369857016Subjects--Topical Terms:

555568
Economics.
Index Terms--Genre/Form:

554714
Electronic books.

Three Essays on Realized Volatility Models for High-Frequency Data.
LDR:03533ntm a2200385Ki 4500 001 909880
005 20180426091049.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9781369857016
035 $a (MiAaPQ)AAI10610818
035 $a AAI10610818
040 $a MiAaPQ $b eng $c MiAaPQ
099 $a TUL $f hyy $c available through World Wide Web
100 1 $a Shen, Ji. $3 1180875
245 1 0 $a Three Essays on Realized Volatility Models for High-Frequency Data.
264 0 $c 2017
300 $a 1 online resource (118 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 78-10(E), Section: A.
500 $a Adviser: Denis Pelletier.
502 $a Thesis (Ph.D.) $c North Carolina State University $d 2017.
504 $a Includes bibliographical references
520 $a The dissertation centers on modeling volatility of stock returns with high frequency data. It contains three chapters.
520 $a In Chapter 1, we propose a multivariate realized GARCH model where the conditional variance of daily returns depends on lagged realized measures of volatility computed with high-frequency data. In turn, a measurement equation contemporaneously links the realized measures and the conditional variance. To greatly reduce the number of parameters in the measurement equation, we use adaptive Lasso and the least-angle regression algorithm. For the GARCH dynamic of the conditional variance, we favor a diagonal BEKK structure applied to rotated variables. Our empirical results suggest that the use of variable selection with adaptive Lasso for the measurement equation instead of OLS leads to improved out-of-sample forecasting performances.
520 $a In Chapter 2, we introduced a realized Markov regime-switching BEKK model which generalizes the BEKK model by using two regimes featuring different variance dynamic specifications. Within each regime, we use a BEKK model to govern the variance. The persistence of both regimes yields an extra source of volatility persistence compared with standard, single-regime BEKK, thereby enhancing the flexibility in describing the volatility persistence of shocks. To reduce the parameters in estimation, we implement covariance targeting with rotated variables. The empirical evidence shows that the regime-switching BEKK model has a significantly better performance in both in-sample fit and out-of-sample forecasting.
520 $a In Chapter 3, we generalize the realized Markov regime-switching GARCH model in Chapter 2 by allowing the regime transition probability to vary over time using pre-determined observable variables which is the realized kernel in this case. This model has more flexibility regarding volatility persistence compared with the one in Chapter 2. To avoid the computational difficulty due to an increasing number of parameters, we utilize this model with one stock only. Our empirical research shows that the use of time-varying transition probabilities leads to better in-sample fit and improved forecasting ability in the near future (about 10 days).
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Economics. $3 555568
650 4 $a Finance. $3 559073
650 4 $a Statistics. $3 556824
655 7 $a Electronic books. $2 local $3 554714
690 $a 0501
690 $a 0508
690 $a 0463
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a North Carolina State University. $b Economics. $3 1180876
773 0 $t Dissertation Abstracts International $g 78-10A(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10610818 $z click for full text (PQDT)