A copula-based markov switching seemingly unrelated regression approach for analysis the demand and supply on sugar market
© Springer International Publishing AG 2016. This paper conducted a Markov switching seemingly unrelated regression without assuming a normal distribution of the error term. We proposed the use of both Archimedean and Elliptical copula classes to join the different marginal of the system equations....
Saved in:
| Main Authors: | , , , |
|---|---|
| 格式: | Book Series |
| 出版: |
2018
|
| 主題: | |
| 在線閱讀: | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85005990012&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/55566 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 總結: | © Springer International Publishing AG 2016. This paper conducted a Markov switching seemingly unrelated regression without assuming a normal distribution of the error term. We proposed the use of both Archimedean and Elliptical copula classes to join the different marginal of the system equations. The results show that normal distribution for both demand and supply equations and joint distribution by Frank copulas present the lowest AIC and BIC. Moreover, the model is, then, applied for estimating the demand and supply in Thai sugar market. Thai export price and Brazil’s export price were found to be the factors affecting the demand and supply of the Thai sugar market. Finally, the results on smoothed probabilities indicate the oversupply condition in Thai sugar market along our sample period. |
|---|