High-order generalized maximum entropy estimator in kink regression model
© 2019 by the Mathematical Association of Thailand. All rights reserved. Investigation was made on the performance of the high-order Generalized Maximum Entropy (GME) estimators, namely Rényi and Tsallis GME, in the nonlinear kink regression context with an aim to replace the Shannon entropy measure...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Journal |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85068475617&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/65688 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Chiang Mai University |
| Summary: | © 2019 by the Mathematical Association of Thailand. All rights reserved. Investigation was made on the performance of the high-order Generalized Maximum Entropy (GME) estimators, namely Rényi and Tsallis GME, in the nonlinear kink regression context with an aim to replace the Shannon entropy measure. Used for performance comparison was the Monte Carlo Simulation to generate the sample size n = 20 and n = 50 with various error distributions. Then, the obtained model was applied to the real data. The results demonstrate that the high-order GME estimators are not much different from the Shannon GME estimator and are not completely superior to the Shannon GME in the simulation study. Nevertheless, according to the MAE criteria, Rényi and Tsallis GME perform better than the Shannon GME. Thus, it can be concluded that high-order GME estimator can be used as alternative tool in the nonlinear econometric framework. |
|---|