论文部分内容阅读
研究中国股票市场中的两个重要指标:股票价格与交易量,随机波动模型具有长期波动性预测能力,只是由于参数估计的困难而没有受到重视.随着马尔可夫链蒙特卡罗(MCMC)方法和计算机计算能力的提高,这种困难是可以克服的.采用基于马尔可夫链蒙特卡罗(MCMC)模拟技术的贝叶斯估计方法,在基于引入预期交易量和非预期交易量的随机波动模型下,对模型参数进行后验分布的构造,并以2005年1月-2012年5月的上证综合指数的每日收盘指数及相应的日成交量序列为样本,通过实证仿真得到参数结果值.结果表明,非预期交易量对股市价格的影响要大于预期交易量.
Studying two important indexes in Chinese stock market: the stock price and the trading volume, the stochastic volatility model has long-term volatility prediction ability, which is not taken seriously because of the difficulty of parameter estimation.With the progress of Markov chain Monte Carlo (MCMC) Method and computer calculation ability, this kind of difficulty can be overcome.Based on the Bayesian estimation method based on Markov chain Monte Carlo (MCMC) simulation technology, based on the introduction of expected and unanticipated trading volume of random Under the volatility model, we construct the posterior distribution of the model parameters. Taking the daily closing index and corresponding daily trading volume of the Shanghai Composite Index from January 2005 to May 2012 as samples, we obtain the parameter results through empirical simulation The results show that the impact of unanticipated trading volume on the stock market price is greater than the expected trading volume.