泊松过程单变点模型的贝叶斯参数估计

Bayesian parameter estimation of Poisson process model with single change-point

利用泊松分布和二项分布的关系,通过引入潜在变量得到了泊松过程单变点模型比较简单的似然函数.得到了未知参数的满条件分布,对满条件分布进行了Gibbs抽样,基于Gibbs样本对参数进行估计.随机模拟试验的结果表明贝叶斯估计的精度较高.

By introducing a latent variable, the simple likelihood function of Poisson process with a change-point is obtained according to the relationship of Poisson distribution to binomial distribution. Then the full conditional distributions of unknown parameters are obtained, Gibbs sampling is performed for them, and the parameters are estimated based on the Gibbs sample. The result of random simulation test shows that the accuracy of Bayesian estimation is comparatively high.