GARCH模型和ECM模型对沪深两市预测的比较分析

GARCH模型反映了经济变量之间特殊的不确定形式：方差随时间变化，所以在金融市场的预测和决策有着重要的作用。鉴于股票和房地产这两个重要的经济指标，本文选择上海和深圳两地的股票收益率的波动性作为研究对象，建立了GARCH模型及其主要变化形式，结果表明基于T分布的GARCH（1，1）模型更好模拟了实际值。另外，本文还对上海，深圳的股市进行协整分析，建立了相应的误差修正模型并对其预测，预测效果比基于T分布的GARCH（1，1）模型更好。

The Comparison and Analysis of GARCH Model and the ECM Model in Predicting Shanghai Stock Index and Shenzhen Stock Index

GARCH model reflects a special feature of economic variables：time-varying variances. So it plays the important role in the financial market. In consideration of the two important economic indexes--the stock market and real estate, we choose the volatility of Shanghai and Shenzhen stocks as the research objects. Then we introduce the form and classes of GARCH in this article. The result indicates that the GARCH （1,1）-T model is better than others. In addition, this paper also conducts the cointegration analysis and establishes the error correction model. And this model shows to be more practical to improve the prediction precision.