[1]刘思杨,蔡 艳*.应用Stan软件包实现IRT模型的贝叶斯参数估计[J].江西师范大学学报(自然科学版),2020,(03):282-291.[doi:10.16357/j.cnki.issn1000-5862.2020.03.12]
　LIU Siyang,CAI Yan*.Using Stan to Implement Bayesian Parameter Estimation of IRT Models[J].Journal of Jiangxi Normal University:Natural Science Edition,2020,(03):282-291.[doi:10.16357/j.cnki.issn1000-5862.2020.03.12]
点击复制

应用Stan软件包实现IRT模型的贝叶斯参数估计()
分享到：

《江西师范大学学报》（自然科学版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2020年03期
页码:
282-291
栏目:
信息科学与技术
出版日期:
2020-06-10

文章信息/Info

Title:
Using Stan to Implement Bayesian Parameter Estimation of IRT Models
文章编号:
1000-5862(2020)03-0282-10
作者:
刘思杨; 蔡 艳*
江西师范大学心理学院,江西 南昌 330022
Author(s):
LIU Siyang; CAI Yan*
College of Psychology,Jiangxi Normal University,Nanchang Jiangxi 30022,China
关键词:
项目反应理论;  汉密尔顿蒙特卡罗算法;  贝叶斯估计;  Stan
Keywords:
item response theory;  HMC;  Bayesian estimation;  Stan
分类号:
B 841
DOI:
10.16357/j.cnki.issn1000-5862.2020.03.12
文献标志码:
A
摘要:
Stan是一个新的用于估计指定统计模型的概率编程语言,它使用了强大而高效的汉密尔顿蒙特卡罗(Hamiltonian Monte Carlo,HMC)抽样算法,相比较传统的Gibbs抽样和Metropolis算法具有显著的效率提升.R软件包“rstan” 链接了R与Stan 2个软件,使得Stan可以借助R的计算环境运行.首先,该文通过3参数Logistic(3PL)模型代码介绍了Stan的程序语言; 其次,该文使用Stan计算2个评估模型-数据拟合的全新指标WAIC和LOO,为应用Stan进行IRT模型相关研究提供了有效的参考工具; 最后,该文还采用了2个真实数据分别考察了Stan在单维IRT模型和多维IRT模型参数估计中的运行表现.研究结果表明:采用一个新的贝叶斯统计软件Stan,通过2个实证研究验证了该方法的有效性与可行性,为国内学者应用Stan进行IRT模型相关研究提供了有效的参考资料.
Abstract:
Stan,a new probabilistic programming language for specifying statistical models,implements the powerful and efficient Hamiltonian Monte Carlo(HMC)sampling algorithm,which is significantly more efficient than the traditional Gibbs sampling and Metropolis algorithms.R package "rstan" links R and Stan,enabling Stan to run with R environment.First,this article introduces the programming language of Stan through the three-parameter logistic(3PL)model code.Secondly,this paper administrates Stan to calculate two new criteria of model-data fitting:WAIC and LOO,which provides an effective reference for IRT model studies.Finally,two types of real data are performed to investigate the performance of Stan in parameter estimation of one-dimensional IRT model and multidimensional IRT model respectively.In conclusion,this paper utilizes a new Bayesian statistical software to estimate the effectiveness and feasibility of this method through two empirical studies,which provides effective references for domestic scholars to apply Stan in IRT model research.

参考文献/References:

[1] Carpenter B,Gelman A,Hoffman M D,et al.Stan:a probabilistic programming language[J].Journal of Statistical Software,2017,76(1):1259-1270.
[2] Kruschke J.Doing Bayesian data analysis:a tutorial with R,JAGS,and Stan[M].Pittsburgh:Academic Press,2014.
[3] Lunn D J,Thomas A,Best N,et al.WinBUGS a Bayesian modelling framework:concepts,structure,and extensibility[J].Statistics and Computing,2000,10(4):325-337.
[4] Lunn D,Jackson C,Best N,et al.The BUGS book:a practical introduction to Bayesian analysis[M].Chapman:Chapman and Hall/CRC,2012.
[5] Plummer M.JAGS:a program for analysis of Bayesian graphical models using Gibbs sampling[EB/OL].[2019-10-18].http://core.ac.UR/display/20935350.
[6] Martin A D,Quinn K M,Park J H.MCMCpack:Markov chain monte carlo in R[EB/OL].[2019-11-13].https://www.oalib.com/paper/2884490#.XtoYuOSHodU.
[7] Geman S,Geman D.Stochastic relaxation,Gibbs distributions,and the Bayesian restoration of images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1984(6):721-741.
[8] Metropolis N,Rosenbluth A W,Rosenbluth M N,et al.Equation of state calculations by fast computing machines[J].The Journal of Chemical Physics,1953,21(6):1087-1092.
[9] Neal R M.Probabilistic inference using Markov chain Monte Carlo methods[D].Toronto,Ontario,Canada:University of Toronto,1993.
[10] Liang Zhongyao,Yu Yanhong,Wang Liqian,et al.A Bayesian ANOVA method to identify the temporal and seasonal dynamics of lake water quality variables[J].Acta Scientiae Circumstantiae,2017,37(11):4170-4177.
[11] Neal R M.MCMC using Hamiltonian dynamics[J].Handbook of Markov Chain Monte Carlo,2011,2(11):113-162.
[12] Banerjee S,Carlin B P,Gelfand A E.Hierarchical modeling and analysis for spatial data[M].Chapmann:Chapman and Hall/CRC,2014.
[13] Stan Development Team.ShinyStan:interactive visual and numerical diagnostics and posterior analysis for Bayesian models[EB/OL].[2019-11-13].https://xueshu.baidu.com/usercenter/paper/show?paperid=652a67f6b2a3f9c
616b0db8ab2d45280&site=xueshu_se.
[14] Stan Development Team.Stan modeling language users guide and reference manual[EB/OL].[2019-11-13].http://citeseerx.ist.psu.edu/viewdoc/download; jsessionid=4605C5A369842DE403C89BB7293BE5D8?doi=10.1.1.372.3101&rep=rep1&type=pdf.
[15] 涂冬波,蔡艳,戴海琦,等.多维项目反应理论:参数估计及其在心理测验中的应用[J].心理学报,2011,43(11):1329-1340.
[16] Curtis S M K.BUGS code for item response theory[J].Journal of Statistical Software,2010,36(1):1-34.
[17] Stone C A,Zhu Xiaowen.Bayesian analysis of item response theory models using SAS[EB/OL].[2019-11-13].https://www.sas.com/content/dam/SAS/support/en/books/bayesian-analysis-of-item-response-theory-models-using-sas/67262_excerpt.pdf.
[18] 刘晋,汪秀琴,李天萍,等.贝叶斯统计分析的新工具:Stan[J].中国卫生统计,2019,36(3):462-465,470.
[19] Rusch T,Maier M J,Hatzinger R.Linear logistic models with relaxed assumptions in R[EB/OL].[2019-11-13].https://link.springer.com/chapter/10.1007%2F978-3-319-00035-0_34.
[20] Team S D.RStan:the R interface to Stan[EB/OL].[2019-11-13].http://ftp.ps.pl/dsk0/CRAN/web/packages/rstan/vignettes/rstan_vignette.pdf.
[21] Swaminathan H,Gifford J A.Bayesian estimation in the three-parameter logistic model[J].Psychometrika,1986,51(4):589-601.
[22] Vehtari A,Gelman A,Gabry J.Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC[J].Statistics and Computing,2017,27(5):1413-1432.
[23] Watanabe S.Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory[J].Journal of Machine Learning Research,2010,11:3571-3594.
[24] Spiegelhalter D,Best N G,Carlin B P,et al.Bayesian measures of model complexity and fit[J].Quality Control and Applied Statistics,2003,48(4):431-432.
[25] Akaike H.A new look at the statistical model identification[M].New York:Springer,1974:215-222.
[26] Schwarz G.Estimating the dimension of a model[J].The Annals of Statistics,1978,6(2):461-464.
[27] Luo Yong,Al Harbi K.Performances of LOO and WAIC as IRT model selection methods[J].Psychological Test and Assessment Modeling,2017,59(2):183.
[28] 蔡华俭,朱臻雯,杨治良.心理类型量表(MBTI)的修订初步[J].应用心理学,2001,7(2):33-37.
[29] Brooks S P,Gelman A.General methods for monitoring convergence of iterative simulations[J].Journal of Computational and Graphical Statistics,1998,7(4):434-455.
[30] Carpenter P A,Just M A,Shell P.What one intelligence test measures:a theoretical account of the processing in the Raven Progressive Matrices Test[J].Psychological Review,1990,97(3):404.
[31] Reckase M D,McKinley R L.Some latent trait theory in a multidimensional latent space[EB/OL].[2019-11-13].https://files.eric.ed.gov/fulltext/ED264265.pdf.
[32] Cowles M K,Carlin B P.Markov chain Monte Carlo convergence diagnostics:a comparative review[J].Journal of the American Statistical Association,1996,91(434):883-904.
[33] Sinharay S.Experiences with Markov chain Monte Carlo convergence assessment in two psychometric examples[J].Journal of Educational and Behavioral Statistics,2004,29(4):461-488.

相似文献/References:

[1]关潮辉,丁树良.基于IRT模型的BP神经网络参数估计的进一步研究[J].江西师范大学学报(自然科学版),2014,(04):434.
　GUAN Chao-hui,DING Shu-liang.Further Research on the Paraneter Estination Method of BP Neural Network Based on Iten ResPonse Theory[J].Journal of Jiangxi Normal University:Natural Science Edition,2014,(03):434.
[2]简小珠,戴海琦.4参数GRM对猜测现象和失误现象的纠正[J].江西师范大学学报(自然科学版),2016,40(02):116.
　JIAN Xiaozhu,DAI Haiqi.Four-Parameter GRM and the Countermeasure to Sleeping and Guessing Phenomena[J].Journal of Jiangxi Normal University:Natural Science Edition,2016,40(03):116.
[3]黎光明,王小婷.非等组锚题设计下IRT等值方法比较及其应用[J].江西师范大学学报(自然科学版),2017,(05):454.
　LI Guangming,WANG Xiaoting.The Comparison of Equating Methods in Non-Equivalent Group with Anchor Test Design Based on Item Response Theory and Its Application[J].Journal of Jiangxi Normal University:Natural Science Edition,2017,(03):454.
[4]戴步云,谢宇欣,简小珠.基于计算机自适应的大学生社会责任感测验[J].江西师范大学学报(自然科学版),2020,(02):142.[doi:10.16357/j.cnki.issn1000-5862.2020.02.05]
　DAI Buyun,XIE Yuxin,JIAN Xiaozhu.The College Students' Social Responsibility Testing Based on Computerized Adaptation[J].Journal of Jiangxi Normal University:Natural Science Edition,2020,(03):142.[doi:10.16357/j.cnki.issn1000-5862.2020.02.05]
[5]张玉柳,赵 波*,陶金洪.基于模糊认知诊断模型的学生认知状态研究[J].江西师范大学学报(自然科学版),2021,(05):452.[doi:10.16357/j.cnki.issn1000-5862.2021.05.02]
　ZHANG Yuliu,ZHAO Bo*,TAO Jinhong.The Study on Students’ Cognitive State Based on Fuzzy Cognitive Diagnosis Framework[J].Journal of Jiangxi Normal University:Natural Science Edition,2021,(03):452.[doi:10.16357/j.cnki.issn1000-5862.2021.05.02]
[6]曾 光,张玉玲,谢晓尧,等.一种改进的4参数等级反应模型和应用[J].江西师范大学学报(自然科学版),2023,(02):124.[doi:10.16357/j.cnki.issn1000-5862.2023.02.02]
　ZENG Guang,ZHANG Yuling,XIE Xiaoyao,et al.The Improved Four-Parameter GRM and Its Application[J].Journal of Jiangxi Normal University:Natural Science Edition,2023,(03):124.[doi:10.16357/j.cnki.issn1000-5862.2023.02.02]

备注/Memo

备注/Memo:
收稿日期:2019-12-16
基金项目:国家自然科学基金(31760288)资助项目.
通信作者:蔡 艳(1979-),女,江西宜春人,教授,博士生导师,主要从事心理统计与测量研究.E-mail:cy1979123@aliyun.com
更新日期/Last Update: 2020-06-10