Welcome Tp CRETA! Contact Us at : +886.2.3366.1072

Welcome To CRETA

國立臺灣大學計量理論與應用研究中心 - CRETA



2018 年 4 月 27 日 CRETA Seminar


2018 年 4 月 27 日 CRETA Seminar


國立臺灣大學計量理論與應用研究中心 (CRETA)、國立臺灣大學財務金融學系及臺灣經濟計量學會 (TES) 將於 2018 年 4 月 27 日舉辦 CRETA Seminar。相關資訊如下。

4 27  CRETA Seminar

日期:2018 年 4 月 27 日 (週五) 下午2:00~5:00

地點:國立臺灣大學管理學院二號館三樓 304 教室



Forecast Accuracy Tests, Consistent Loss Functions and Relevant Applications


  The first part of the presentation is based on the paper “Testing Forecast Accuracy of Expectiles and Quantiles with the Extremal Consistent Loss Functions”. In this paper, we develop statistical tests for comparing performances of forecasting expectiles and quantiles of a random variable under consistent loss functions. The test statistics are constructed with the extremal consistent loss functions of Ehm et al. (2016). The null hypothesis of the tests is that a benchmark forecast at least performs equally well as a competitive one under all extremal consistent loss functions. It can be shown that if such a null holds, the benchmark will also perform at least equally well as the competitor under all consistent loss functions. We discuss asymptotic properties of the proposed test statistics and propose to use the re-centered bootstrap to construct their empirical distributions. Through simulations, we show the proposed test statistics perform reasonably well. We then apply the proposed method on (1) re-examining abilities of some often-used predictors on forecasting risk premium of the S&P500 index; (2) comparing performances of experts' forecasts on annual growth of U.S. real gross domestic product; (3) evaluating performances of estimated daily value at risk of the S&P500 index.


  The second part of the presentation is about applications of some newly proposed consistent loss functions. We will introduce how to estimate Value-at-Risk (VaR) and expected shortfall (ES) with a consistent loss function proposed by Fissler and Ziegel (2016). With the FZ loss function, we propose a two-step method to conduct the estimations. We then estimate various models and compare their performances of in-sample estimations and out-of-sample forecasts. We find models added with realized volatility as the exogenous variable perform better. Other possible applications will also be discussed.






為方便場地安排及人數預估,欲參加CRETA Seminar的朋友們,煩請事先報名。




             其他參加者報名費為NT$200 (當天將開放現場繳交台灣經濟計量學會 2018 年年度會費)。

報名期限:2018/4/26 (四) 13:30


為方便臺灣經濟計量學會 (TES) 會員繳納 2018 年度會費,本次活動開放現場繳納會費,亦歡迎大家介紹非會員朋友加入 TES。更多研討會資訊請見 TES 網站:http://www.tesociety.org.tw/main.php


議  程:

下午 1:30 ~ 2:00 報到

下午 2:00 ~ 3:20 First session

下午 3:20 ~ 3:40 Tea Break 

下午 3:40 ~ 5:00 Second session