The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
577
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
|
Sponsor:
|
International Chinese Statistical Association
|
Abstract - #304549 |
Title:
|
Multivariate Density Forecast: An Application to Multivariate Duration Models
|
Author(s):
|
Yingying Sun*+ and Gloria González-Rivera
|
Companies:
|
University of California at Riverside and University of California at Riverside
|
Address:
|
875 Grape Street, Riverside, CA, 92507, United States
|
Keywords:
|
autocontour ;
multivariate duration ;
copula ;
Probability integral transform ;
density forecast
|
Abstract:
|
We test the null hypothesis of a correct density forecast by applying the autocontour methodology (González-Rivera et al. 2011) on i.i.d. U[0,1] probability integral transforms. Our testing is directly applicable to univariate and multivariate density models with either continuous or discrete random variables, and it allows the researcher to focus on different areas of the conditional density model to assess those regions of interest. We construct hyper-cubes of different sizes within the maximum hyper-cube formed by a multidimensional uniform density [0,1]^n. We assess the location of the empirical PITs within the corresponding population hyper-cubes. If the density model is correct, the volumes of the population hyper-cubes must be the same as those in their empirical counterparts. We propose several statistical tests with standard asymptotic distributions and assess multivariate duration models. We choose a copula to produce the multivariate distribution. Our tests reject the Gaussian copula for the substantial asymmetry for long versus short duration. Instead, a Clayton copula seems to be preferred when durations are short and the number of transactions is large.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.