頁籤選單縮合
題名 | 動態因子模型對股價預測能力之研究--以亞洲國家之追蹤資料分析=A Dynamic Factor Model for Forecasting Stock Price: A Panel Data Approach in Asian Countries |
---|---|
作者 | 胡育豪; Hu, Yu-hau; |
期刊 | 全球商業經營管理學報 |
出版日期 | 20160900 |
卷期 | 8 2016.09[民105.09] |
頁次 | 頁31-42 |
分類號 | 563.54 |
語文 | chi |
關鍵詞 | 動態因子模型; 追蹤資料; 股價; 樣本外預測; Dynamic factor model; Panel data; Stock price; Out-of-sample forecast; |
中文摘要 | 本篇文章主要是應用追蹤資料(Panel data)的方法,架構動態因子要素模型(dynamic factor model),以進行亞洲國家股價預測,研究期間為1997:1 到2014:4 的股價月資料。參考Engel、Mark 與West(2013)進行匯率預測方法,本研究是將跨國歷史股價分析出影響股價變動的共同因子,此部分可解釋為國際因素、市場風險貼水或市場噪音所構成,也是傳統基本面變數所無法解釋的部分。最後透過一個月、三個月、六個月、一年期及一年半等不同的預測期間,觀察動態因子要素模型之預測能力與隨機遊走模型和傳統基本面變數在樣本外預測能力的差異。 |
英文摘要 | This paper uses the dynamic factor model in order to investigate the performance of forecasting stock price in Asian countries. The panel data approach is being employed covering monthly data from 1997:1 to 2014:4. According to the methodology of Engel, Mark and West (2013) in forecasting exchange rate, we try to analyze the impact of common factor in cross-country historical stock prices. The common factor could be interpreted as part of the international factors, the risk premium or the market noise, which seem not be explained by traditional fundamental variables. Through the different forecasting period from one month to one year, we can observe the out-of-sample forecasting performance among random walk model、traditional fundamental model and dynamic factor model. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。