基于bayes方法的非参数估计-概率论与数理统计专业论文.docxVIP

摘 摘 要 在非参数估计的研究中，Bayesian方法是一种常用也非常有效的手段．在本 文中。我们应用Bayesian的思想解决了一类非参数的估计的问题：有界区间上的变 带宽局部多项式估计和时间序列趋势项的非参数估计．对于第一个问题， Fan and Gijbels(1996)提出了变带宽局部多项式估计的思想：把区间依据数据容量划分成一 些小的区间，在每个子区间上用局部多项式来进行估计并根据某些适当的准则来确 定其带宽，在本文中，我们把要划分的小区间的个数和区间端点都作为随机变量， 然后利用可逆跳跃的马尔柯夫蒙特卡洛(RJMCMC)的方法来解决这一问题．对于 第二个问题，我们从尺度空问(scale space)的观点来考虑带AR(p)误差的时间序列 的趋势函数的多尺度贝叶斯分析，目p把趋势函数的核估计中的带宽(h)看成是随机 参数，并服从混合r先验或者混合对数正态的先验，通过对h的贝叶斯分析来分析 趋势函数的长期、中期、短期的情况．为从后验分布中抽取样本，我们应用Gibbs 抽样算法得到Markov-chain样本．最后我们给出了数值例子和总结． 关键词：带宽 RJMCMC Gibbs采样Dirichlet分布 混合分布 数据扩大 AbstractIn Abstract In the field of nonpaxametrlc estimation．Bayesian method is not only extensively used but alsovery effective meaEts，In this paper，we 3pplled Bayesian idea to address kind of nonparametric estimation problem：local polynomial estimation with variable bandwidth bounded interval and the trend function of time series of uonpaJametrie estimation．For the first problem，Fan and Gijbels(1996)advise idea of local poly- nomiM estimation with variable bandwidth：partition the interval into some subintervals according to the data size，on each subinterval applied local polynomial estimation method and determined the bandwidth according to some suitable criterions．In this paper，we regarded both the number of suhintervai and the end point of subinterval random parameters，then applied reversibl争jump MCMC methods to this problem．For the ond problem，we considered the multiscale Bayesian analysis of trend function of time series with AR(p)errors from the point of scale space．That is to 8a*we regarded the bandwidth(h1 in the trend function’S kernel estimation random parameter．Through the Bayesian analysis of h，we analysed the trend function’s long-term，middle-term and short·term ca．ses，In order to sampling的m posterior distrubution．we applied the Gibbs sa．mpling algorithm to this problem and obtained the Markov—chain samples．In the end， we given some simulated examples and summarization． Keywords：bandwidth，tLIMCMC，Gibbs sampting，Dirichlet distribution，hy