:::
*Date: Oct.28(Fri.), 2011, 14:00 pm – 16:00 pm
*Venue: Kuan Te Lecture Hall, 2F, College of Management, NTU
(台灣大學管理學院一號館 2F 冠德講堂)
*Topic:
Optimal Forecasts in the Presence of Structural Breaks
[Regirstration Fee]
台灣大學在學學生及現任教職員和台灣經濟計量學會會員為免費參加
其他參加者報名費為 NT$600
(當天將開放現場繳交台灣經濟計量學會 2012 年年度會費)
[About the Speaker]
Professor M. Hashem Pesaran is a one of the most prestigious economist to the world, who is currently Professor at the Faculty of Economics at the University of Cambridge and a Professorial Fellow of Trinity College, Cambridge. He holds the fellowship of the British Academy, the Econometric Society, and also the Journal of Econometrics.
Prof. Pesaran has over 130 publications in leading scientific journals in the areas of econometrics, empirical macroeconomics and the Iranian economy, and is an expert in the economics of Oil, the Middle East, and other emerging regions.
Prof. Pesaran’s diligent effort and extraordinary insight have been recognized with honors such as the 1992 Royal Economic Society Prize for the best article and the Econometric Reviews Best Paper Award 2002-2004.
Besides the profession experience of head of the Economic Research Department of the Central Bank of Iran and the Director of the Applied Econometrics Program at UCLA, Prof. Pesaran devotes much in academic enhancement by participating in the founding of the Journal of Applied Econometrics and by co-developing Microfit, a widely-utilized econometric software package.
[Lecture Overview]
The paper considers alternative approaches to taking account of structural breaks in forecasting, and derives weights for observations in weighted least square regressions that lead to optimal forecasts in the presence of breaks. Analytical expressions are provided in the case of a single regressor. For more complicated models asympotically valid weights are provided as well as exact weights that can be computed numerically. In the case of a single break it is shown that the value of the optimal weight is the same across observations within a given regime and differs only across the regimes. The analysis is also generalized to the case of multiple breaks. The forecasts based on optimal weights are shown to be superior to forecasts from post-break estimation windows, optimal estimation windows, averaging across estimation windows and exponential smoothing. In practice where information on structural breaks is uncertain a robust optimal weight forecasting procedure is proposed, and a number of empirical applications are provided.
[Program]
Oct.28 (Fri.) Kuan Te Lecture Hall, 2F (二樓冠德講堂)
13:30-14:00: Registration
14:00-15:30: Lecture
15:30-16:00: Tea Time and Discussion
*Lectures in English
會議活動
- 標題：CRETA Workshop on Advanced Econometrics 11_Professor M. Hashem Pesaran
- 公告日期：2011-10-13
國立臺灣大學計量理論與應用研究中心 (CRETA) 很榮幸邀請到 University of Cambridge 的 Professor M. Hashem Pesaran 於 10 月 28 日(五)至本中心訪問，並於 CRETA Workshop on Advanced Econometrics 11 進行專題演講，演講主題為 Optimal Forecasts in the Presence of Structural Breaks。
Professor M. Hashem Pesaran 為全球計量領域知名的學者，此行來訪及學術分享必定帶給與會者相當大的收穫！
專題演講地點為台大管理學院一號館 2F 冠德講堂。歡迎大家踴躍報名參加。欲參加者，請於 10 月 26 日(三)中午前至 CRETA 網站線上報名 <http://www.creta.org.tw/events/view/38 > 。
Professor M. Hashem Pesaran 為全球計量領域知名的學者，此行來訪及學術分享必定帶給與會者相當大的收穫！
專題演講地點為台大管理學院一號館 2F 冠德講堂。歡迎大家踴躍報名參加。欲參加者，請於 10 月 26 日(三)中午前至 CRETA 網站線上報名 <http://www.creta.org.tw/events/view/38 > 。
*Date: Oct.28(Fri.), 2011, 14:00 pm – 16:00 pm
*Venue: Kuan Te Lecture Hall, 2F, College of Management, NTU
(台灣大學管理學院一號館 2F 冠德講堂)
*Topic:
Optimal Forecasts in the Presence of Structural Breaks
[Regirstration Fee]
台灣大學在學學生及現任教職員和台灣經濟計量學會會員為免費參加
其他參加者報名費為 NT$600
(當天將開放現場繳交台灣經濟計量學會 2012 年年度會費)
[About the Speaker]
Professor M. Hashem Pesaran is a one of the most prestigious economist to the world, who is currently Professor at the Faculty of Economics at the University of Cambridge and a Professorial Fellow of Trinity College, Cambridge. He holds the fellowship of the British Academy, the Econometric Society, and also the Journal of Econometrics.
Prof. Pesaran has over 130 publications in leading scientific journals in the areas of econometrics, empirical macroeconomics and the Iranian economy, and is an expert in the economics of Oil, the Middle East, and other emerging regions.
Prof. Pesaran’s diligent effort and extraordinary insight have been recognized with honors such as the 1992 Royal Economic Society Prize for the best article and the Econometric Reviews Best Paper Award 2002-2004.
Besides the profession experience of head of the Economic Research Department of the Central Bank of Iran and the Director of the Applied Econometrics Program at UCLA, Prof. Pesaran devotes much in academic enhancement by participating in the founding of the Journal of Applied Econometrics and by co-developing Microfit, a widely-utilized econometric software package.
[Lecture Overview]
The paper considers alternative approaches to taking account of structural breaks in forecasting, and derives weights for observations in weighted least square regressions that lead to optimal forecasts in the presence of breaks. Analytical expressions are provided in the case of a single regressor. For more complicated models asympotically valid weights are provided as well as exact weights that can be computed numerically. In the case of a single break it is shown that the value of the optimal weight is the same across observations within a given regime and differs only across the regimes. The analysis is also generalized to the case of multiple breaks. The forecasts based on optimal weights are shown to be superior to forecasts from post-break estimation windows, optimal estimation windows, averaging across estimation windows and exponential smoothing. In practice where information on structural breaks is uncertain a robust optimal weight forecasting procedure is proposed, and a number of empirical applications are provided.
[Program]
Oct.28 (Fri.) Kuan Te Lecture Hall, 2F (二樓冠德講堂)
13:30-14:00: Registration
14:00-15:30: Lecture
15:30-16:00: Tea Time and Discussion
*Lectures in English
- 附件檔案：
- 參考連結：http://www.creta.org.tw/events/view/38
- 張貼人：網站管理員
- 最後修改時間：2011-10-13 PM 12:52
