摘要
目的探讨匹配的病例对照研究的Bayes方法。方法构建病例对照研究的Bayes多项式分布模型,利用WinBUGS软件用Gibbs抽样的MCMC模拟技术估计疾病与暴露之间的优势比。结果通过WinBUGS软件实例分析,所得结果与经典方法相比,认为两者结果相似,但是部分区间较宽。结论病例对照研究的Bayes模型的解释合理、恰当。
Objective To explore matched case - control studies using Bayes method. Methods Constructing Bayesian polynomial model for the data of case - control study and using the software WinBUGS. The odds ratio between disease and expose based on Bayesian theory was made through MCMC using Gibbs sampling. Results An example was analyzed by WinBUGS, and the result was the same as the classical method, but having wider interval, Conclusion The Bayesian method in matched case - control study was reasonable and appropriate.
出处
《菏泽医学专科学校学报》
2007年第3期13-15,共3页
Journal of Heze Medical College