|
Activity Number:
|
294
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #308275 |
|
Title:
|
Prediction Interval Estimation in Transformed ARMA Models
|
|
Author(s):
|
Hyemin Cho*+ and Sungun Oh and In-Kwon Yeo
|
|
Companies:
|
Sookmyung Women's University and Sookmyung Women's University and Sookmyung Women's University
|
|
Address:
|
Chungpa-don, Yongsan-gu, Seoul, 140-742, South Korea
|
|
Keywords:
|
ARMA models ; Coverage probability ; Prediction interval ; Smearing estimation ; Yeo-Johnson transformation
|
|
Abstract:
|
One of main aspects of time series analysis is to forecast future values of series based on observations up to a given time. The prediction interval for future values is usually obtained under the normality assumption. When the assumption is seriously violated, a transformation of data may permit the valid use of the normal theory. We investigate the prediction problem for future values in the original scale when transformations are applied in ARMA models. In this study, Yeo-Johnson transformation which can solve the skewness of data is applied and the smearing estimation is employed to reduce the bias. We also estimate the prediction intervals. We present conditions that ensure that the intervals have asymptotically correct coverage probability. Simulations show that the coverage probabilities of proposed intervals are closer to the nominal level than those of usual intervals.
|
