We integrate relevant macroeconomic and financial fundamentals into an aligned index to predict bear markets. The partial least squares approach allows us to eliminate the common noise component in fundamental variables we consider. Using the Markov regime-switching model to date the bear market based on the S&P 500 index returns, we find the aligned predictor performs better in forecasting the bear market than any individual variable composed into the aligned predictor. We further provide evidence that the aligned fundamental index outperforms the aligned sentiments index used in Huang et al. (2015) in both in-sample and out-of-sample forecasts of bear markets.
本文以總經與財務變數建構預測熊市之統整指標,使用偏最小平方(partial least squares, PLS)法,移除眾多變數的共同雜訊成分。透過馬可夫狀態轉換模型(Markov regime-switching model)衡量美國S&P500指數之熊市發生的機率,我們發現PLS統整指標預測熊市發生機率的表現優於個別的總經或財務變數之預測表現,而且其樣本內與樣本外預測能力,皆優於Huang et al.(2015)的投資情緒統整指標。
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。