資料載入中.....
|
請使用永久網址來引用或連結此文件:
https://nccur.lib.nccu.edu.tw/handle/140.119/158283
|
| 題名: | 以具有狀態轉換的高維度因子模型預測台股熊市
A High-Dimensional Factor Model with Regime Switching for Forecasting Bear Markets in the Taiwan Stock Market |
| 作者: | 陳默然
Chen, Mo-Jan |
| 貢獻者: | 徐士勛
Hsu, Shih-Hsun
陳默然
Chen, Mo-Jan |
| 關鍵詞: | 狀態轉換
高維度因子模型
EM演算法
熊市預測
Regime switching
High-dimensional factor model
EM algorithm
Bear market prediction |
| 日期: | 2025 |
| 上傳時間: | 2025-08-04 12:52:11 (UTC+8) |
| 摘要: | 金融市場的牛熊市循環一直是投資者與政策制定者關注的重要議題。過去的市場狀態預測多採用單一或少數經濟指標作為預測變數，難以充分利用現代金融市場的大量可獲取數據。為了處理高維度數據，高維度因子模型受到廣泛應用，其能捕捉大量變數的共同變動，然而當市場狀態轉換等結構性改變發生時，會導致模型的預測準確性下降。
本研究將狀態轉換納入因子模型以處理結構性改變，參考 Urga and Wang (2024) 所提出之具有狀態轉換 (regime switching) 的高維度因子模型 (factor model)，使用EM演算法 (Expectation-Maximization Algorithm) 迭代估計狀態機率與因子負荷，並聯合估計參數。
本文分別使用台股股價資料捕捉牛市與熊市狀態，用於預測熊市發生機率；以及使用股價報酬率資料捕捉短期的高報酬與低報酬狀態，用於建構市場擇時策略。實證結果顯示，儘管模型具有良好的模擬效果，但在樣本內與樣本外預測階段對實際熊市狀態的預測效果有限。然而，基於股價報酬率狀態轉換訊號所建構的市場擇時策略在投資績效上優於買入持有策略，展現出潛在的實務應用價值。本研究拓展了高維度因子模型的應用範疇，並為台灣股市研究提供了新的分析視角。
The bull and bear market cycles in financial markets have long been a major focus for investors and policymakers. Previous market state predictions have mostly adopted single or few economic indicators as predictor variables, making it difficult to fully utilize the large amount of available data in modern financial markets. To handle high-dimensional data, high-dimensional factor models are widely applied, as they can capture common movements of a large number of variables. However, when structural changes such as market regime switching occur, the predictive accuracy of the model decreases.
This study incorporates regime switching into factor models to handle structural changes, referring to the high-dimensional factor model with regime switching proposed by Urga and Wang (2024). We use the Expectation-Maximization (EM) algorithm to iteratively estimate state probabilities and factor loadings, and jointly estimate parameters.
This paper uses Taiwan stock price data to capture bull and bear market states for forecasting bear market occurrence probabilities, and uses stock return data to capture short-term high return and low return states for constructing market timing strategies. Empirical results show that although the model has good simulation performance, the predictive effectiveness for actual bear market states is limited in both in-sample and out-of-sample prediction stages. However, the market timing strategy constructed based on stock return regime switching signals outperforms the buy-and-hold strategy in investment performance, demonstrating potential practical application value. This study expands the application scope of high-dimensional factor models and provides a new analytical perspective for Taiwan stock market research. |
| 參考文獻: | Bai, J. and Ng, S. (2002). Determining the number of factors in approximate factor models. Econometrica, 70(1):191–221.
Baltagi, B. H., Kao, C., and Wang, F. (2021). Estimating and testing high dimensional factor models with multiple structural changes. Journal of Econometrics, 220(2):349–365.
Banerjee, A., Marcellino, M., and Masten, I. (2008). Forecasting macroeconomic variables using diffusion indexes in short samples with structural change. In Forecasting in the presence of structural breaks and model uncertainty, volume 3, pages 149–194. Emerald Group Publishing Limited.
Bry, G. and Boschan, C. (1971). Programmed selection of cyclical turning points. In Cyclical analysis of time series: Selected procedures and computer programs, pages 7–63. NBER.
Chauvet, M. and Potter, S. (2000). Coincident and leading indicators of the stock market. Journal of Empirical Finance, 7(1):87–111.
Chen, S.-S. (2007). Does monetary policy have asymmetric effects on stock returns? Journal of money, credit and banking, 39(2-3):667–688.
Chen, S.-S. (2009). Predicting the bear stock market: Macroeconomic variables as leading indicators. Journal of Banking & Finance, 33(2):211–223.
Chen, S.-S. (2012). Revisiting the empirical linkages between stock returns and trading volume. Journal of Banking & Finance, 36(6):1781–1788.
Cheng, X., Liao, Z., and Schorfheide, F. (2016). Shrinkage estimation of high-dimensional factor models with structural instabilities. The Review of Economic Studies, 83(4):1511–1543.
Corradi, V. and Swanson, N. R. (2014). Testing for structural stability of factor augmented forecasting models. Journal of Econometrics, 182(1):100–118.
Cunado, J., Gil-Alana, L., and de Gracia, F. P. (2010). Mean reversion in stock market prices: New evidence based on bull and bear markets. Research in International Business and Finance, 24(2):113–122.
Diebold, F. X. and Rudebusch, G. D. (1989). Scoring the Leading Indicators. Journal of Business, 62(3):369–391.
Guidolin, M. and Timmermann, A. (2005). Economic implications of bull and bear regimes in UK stock and bond returns. The Economic Journal, 115(500):111–143.
Hamilton, J. D. (1990). Analysis of time series subject to changes in regime. Journal of econometrics, 45(1-2):39–70.
Harding, D. and Pagan, A. (2003). A comparison of two business cycle dating methods. Journal of Economic Dynamics and Control, 27(9):1681–1690.
Kim, C.-J. (1994). Dynamic linear models with Markov-switching. Journal of econometrics, 60(1-2):1–22.
Kim, M.-J. and Yoo, J.-S. (1995). New index of coincident indicators: A multivariate Markov switching factor model approach. Journal of Monetary Economics, 36(3):607–630.
Kole, E. and Van Dijk, D. (2017). How to identify and forecast bull and bear markets? Journal of Applied Econometrics, 32(1):120–139.
Liu, W., Resnick, B. G., and Shoesmith, G. L. (2004). Market timing of international stock markets using the yield spread. Journal of Financial Research, 27(3):373–391.
Liu, X. and Chen, R. (2016). Regime-switching factor models for high-dimensional time series. Statistica Sinica, pages 1427–1451.
Maheu, J. M. and McCurdy, T. H. (2000). Identifying bull and bear markets in stock returns. Journal of Business & Economic Statistics, 18(1):100–112.
Pagan, A. R. and Sossounov, K. A. (2003). A simple framework for analysing bull and bear markets. Journal of applied econometrics, 18(1):23–46.
Pelger, M. and Xiong, R. (2022). State-varying factor models of large dimensions. Journal of Business & Economic Statistics, 40(3):1315–1333.
Resnick, B. G. and Shoesmith, G. L. (2002). Using the yield curve to time the stock market. Financial Analysts Journal, 58(3):82–90.
Stock, J. H. and Watson, M. W. (1996). Evidence on structural instability in macroeconomic time series relations. Journal of Business & Economic Statistics, 14(1):11–30.
Urga, G. and Wang, F. (2024). Estimation and inference for high dimensional factor model with regime switching. Journal of Econometrics, 241(2):105752.
Wang, M., Lin, Y.-H., and Mikhelson, I. (2020). Regime-switching factor investing with hidden Markov models. Journal of Risk and Financial Management, 13(12):311.
Yamamoto, Y. (2016). Forecasting with nonspurious factors in US macroeconomic time series. Journal of Business & Economic Statistics, 34(1):81–106.
Yu, X., Chen, Z., Xu, W., and Fu, J. (2017). Forecasting bull and bear markets: Evidence from China. Emerging Markets Finance and Trade, 53(8):1720–1733.
李偉銘、吳淑貞與黃啟泰 (2015). 總體經濟變數對臺灣股市之大盤及類股熊市預測表現之探討. 經濟研究, 51(2):171–224.
李春長、梁志民與周幸蓉 (2008). 台灣房地產景氣循環轉折點認定之研究-雙變量馬可夫轉換模型之應用. Journal of Taiwan Land Research, 11(2):155–177. |
| 描述: | 碩士
國立政治大學
經濟學系
112258032 |
| 資料來源: | http://thesis.lib.nccu.edu.tw/record/#G0112258032 |
| 資料類型: | thesis |
| 顯示於類別: | [經濟學系] 學位論文
|
文件中的檔案:
| 檔案 |
描述 |
大小 | 格式 | 瀏覽次數 |
|---|
| 803201.pdf | | 1973Kb | Adobe PDF | 0 | 檢視/開啟 |
|
在政大典藏中所有的資料項目都受到原著作權保護.
|
