State Space Model,Cointegration and Stock Price Forecasting
状态空间模型、协调积分与股票价格预测

This paper investigates the dynamic behavior of five stock prices over the volatile period from July 1996 to December 1996, finding that in the trend model the five series are cointegrated with one dominant common trend (with an estimate root of 0.99). In addition a weak trend and two complex roots are found in the cycle model. These findings are validated by out of sample forecasting and associate statistic creteria. The out of sample forecasts are evaluated nonparametrically in a test due to Henriksson and Merton.