透過您的圖書館登入
IP:3.139.70.131
  • 學位論文

是否有比歷史平均法更有效預測股票市場溢酬的方法?-以美國市場為例

Is there a way to predict the stock market risk premium better than historical average? Evidence from the US market stock market

指導教授 : 顏佑銘
本文將於2025/07/22開放下載。若您希望在開放下載時收到通知,可將文章加入收藏

摘要


本研究乃運用Welch and Goyal(2008)所提出的1945年至2018年的十二項經濟變數資料去預測股票市場溢酬。首先重新檢驗單一變數的樣本內測試(in-sample test)及樣本外測試(out-of-sample test)的結果,接著透過組合變數模型估計,最後再以模型平均法(Model Averaging)建構新的模型。希望藉由新的建構模型方法,比較不同預測模型的預測能力。實證結果發現,以上所建構的模型,皆無法打敗歷史平均法(historical average method)。

並列摘要


In this paper, I use the twelve economic variables of Welch and Goyal (2008) to predict the stock market premium. First of all, I reexamine the in-sample and out-of-sample test. After that, I establish the combination variable to predict the stock market. Finally, I use the model averaging to establish new models. Empirical results display that, the whole models fail to beat the historical average.

參考文獻


中文參考文獻
1.陳旭昇(2009),"時間序列-總體經濟與財務金融與財務金融之應用”,台北:東華書局
英文參考文獻
1.Ang, Andrew Bekaert, Geert. (2007) “Stock return predictability: Is it there? “, Review of Financial Studies, 651-707
2.Baetje, FabianMenkhoff, Lukas.(2016) “Equity premium prediction: Are economic and technical indicators unstable?”, International Journal of Forecasting, 1193-1207

延伸閱讀