Combining Logistic Regression and Markov Chain Monte-Carlo Describe the Relationship between Exposure to a Given Dose of Radiation and its Effect on Clostridium tyrobutyricum Strains
-
摘要: 利用马尔可夫链蒙特卡罗(Markov Chain Monte Carlo,MCMC)方法估计Logistic回归模型中的参数,就是要构造一个以参数的后验分布为其平稳分布的非周期不可约的马尔可夫链,然后用该平稳分布中抽出的样本点计算蒙特卡罗积分。上述理论方法可以解决实验样本数据由于存在定和约束和多重共线性、在进行经典的logistic回归建模时的困难问题。基于此方法,研究了丁酸梭菌株对于给定辐照区间剂量的应答趋势,用模型挖掘数据所隐含的内在信息并导出了Logistic回归模型参数的贝叶斯框架下的50%,90%,95%和99%的置信区间。结果表明,运用Logistic与马尔可夫链耦合模型在有关给定辐射剂量对于微生物作用效果问题的logistic回归建模中具有较大的科学性与很好的使用性,从而可以为辐照诱变处理微生物制定辐照剂量区提供理论支持和回归技术借鉴。
Using the Markov Chain Monte-Carlo method to estimate the parameters in the Logistic regression model, we constructed a non-periodic irreducible Markov Chain with the posterior distribution of the parameters as stationary distribution, and then used the sample points extracted from the stationary distribution to calculate the Monte-Carlo integral. The above theoretical method can solve the difficult problem of classical logistic regression modeling because of the existence and limitation of the experimental sample data and the multicollinearity. In the classical regression setup with a continuous response, the predicted values can range over all real numbers. Therefore, a different modelling technique is needed. In this work, the results describe in detail a previously unknown lethality trend following 12C6+ heavy-ion irradiation of Clostridium tyrobutyricum. By Markov Chain Monte-Carlo can calculate the model fit for a randomly selected subset of the chain and calculate the predictive envelope of the model. The grey areas in the plot correspond to 50%, 90%, 95%, and 99% posterior regions. More importantly, although this study focused on the use of the method in heavy-ion irradiation of microbial, its results are broadly applicable.-
关键词:
- Logistic回归 /
- 马尔可夫链 /
- 辐射剂量 /
- 丁酸梭菌
Abstract: Using the Markov Chain Monte-Carlo method to estimate the parameters in the Logistic regression model, we constructed a non-periodic irreducible Markov Chain with the posterior distribution of the parameters as stationary distribution, and then used the sample points extracted from the stationary distribution to calculate the Monte-Carlo integral. The above theoretical method can solve the difficult problem of classical logistic regression modeling because of the existence and limitation of the experimental sample data and the multicollinearity. In the classical regression setup with a continuous response, the predicted values can range over all real numbers. Therefore, a different modelling technique is needed. In this work, the results describe in detail a previously unknown lethality trend following 12C6+ heavy-ion irradiation of Clostridium tyrobutyricum. By Markov Chain Monte-Carlo can calculate the model fit for a randomly selected subset of the chain and calculate the predictive envelope of the model. The grey areas in the plot correspond to 50%, 90%, 95%, and 99% posterior regions. More importantly, although this study focused on the use of the method in heavy-ion irradiation of microbial, its results are broadly applicable. -
[1] KUTNER M H. Applied Linear Statistical Models[M]. Chicago:Irwin, 1996. 4:318. [2] KUTNER M H, NACHTSHEIM C, NETER J. Applied Linear Regression Models[M]. McGraw-Hill/Irwin, 2004. [3] PEDUZZI P, CONCATO J, KEMPER E, et al. Journal of Clinical Epidemiology, 1996, 49(12):1373. [4] HOSMER J D W, LEMESHOW S. Applied Logistic Regression[M]. John Wiley & Sons,2004. [5] HARRELL F. Regression Modeling Strategies:With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis[M]. Springer, 2015. [6] MENARD S. Applied Logistic Regression Analysis[M]. Sage, 2002. 106. [7] FRIEDMAN J, HASTIE T, TIBSHIRANI R. The Annals of Statistics, 2000, 28(2):337. [8] MOTULSKY H, CHRISTOPOULOS A. Fitting Models to Biological Data Using Linear and Nonlinear Regression:APractical Guide to Curve Fitting[M]. OUP USA, 2004. [9] POWELL J L. Journal of Econometrics, 1984, 25(3):303. [10] MIN C, ZELLNER A. Journal of Econometrics, 1993, 56(1):89. [11] SIMS C A, ZHA T. International Economic Review, 1998:949. [12] GILL J. Bayesian Methods:A Social and Behavioral Sciences Approach[M]. Florida:CRCPress, 2014. 20. [13] GEYER C J. Statistical Science, 1992:473. [14] GREEN P J. Biometrika, 1995, 82(4):711. [15] GILKS W R. Markov Chain Monte Carlo[M]. John Wiley & Sons, 2005. [16] HASTINGS W K.Biometrika, 1970, 57(1):97. [17] NEAL R M. Statistics and Computing, 2001, 11(2):125. [18] HAMMERSLEY J. Monte Carlo Methods[M]. Springer Science & Business Media, 2013. [19] NEAL R M. Journal of Computational and Graphical Statistics, 2000, 9(2):249. [20] METROPOLIS N, ROSENBIUTH A W, ROSENBIUTH M N, et al. The Journal of Chemical Physics, 1953, 21(6):1087. [21] GELFAND A E, HILLS S E, RACINE P A, et al. Journal of The American Statistical Association, 1990, 85(412):972. [22] QURESHI N, MEAGHER M M, HUANG J, et al. Journal of Membrane Science, 2001, 187(1):93. [23] TRAN H T M, CHEIRSILP B, HODGSONod B, et al. Biochemical Engineering Journal, 2010, 48(2):260. [24] ZHOU X, XIN Z J, LU X H, et al. Bioresource Technology, 2013, 137:386. [25] ZHOU X, LU X H, LI X H, et al. Biotechnology for Biofuels, 2014, 7(1):1. [26] KAZAMA Y, SAITO H, YAMAMOTO Y Y, et al. Plant Biotechnology, 2008, 25(1):113. [27] MATUO Y, NISHIJIMA S, HASE Y, et al. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis, 2006, 602(1):7.
计量
- 文章访问数: 1193
- HTML全文浏览量: 144
- PDF下载量: 199
- 被引次数: 0