[1]杨林山,曹亦薇.贝叶斯理论框架下的2种纵向缺失数据处理方法的比较——以潜在变量增长曲线模型为例[J].江西师范大学学报(自然科学版),2012,(05):461-465.
　YANG Lin-shan,CAO Yi-wei.The Comparision of Two Approaches to Bayesian Method for Missing Data in Longitudinal Model —— Growth Curve Model for Example[J].Journal of Jiangxi Normal University:Natural Science Edition,2012,(05):461-465.
点击复制

贝叶斯理论框架下的2种纵向缺失数据处理方法的比较——以潜在变量增长曲线模型为例()
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:
2012年05期
页码:
461-465
栏目:
出版日期:
2012-10-01

文章信息/Info

Title:
The Comparision of Two Approaches to Bayesian Method for Missing Data in Longitudinal Model —— Growth Curve Model for Example
作者:
杨林山;曹亦薇
深圳市海云天教育测评公司, 广东 深圳518067；深圳大学师范学院, 广东 深圳 518060
Author(s):
YANG Lin-shan;  CAO Yi-wei
关键词:
缺失数据; 完全贝叶斯; 部分贝叶斯; 纵向模型; WinBUGS
Keywords:
missing data;  fully Bayesian;  partially Bayesian;  longitudinal models;  WinBUGS
分类号:
B841.2
文献标志码:
A
摘要:
在贝叶斯估计框架下,通过模拟研究比较完全贝叶斯和部分贝叶斯方法对参数估计的影响.研究结果表明:随着缺失比例的增加,2种方法得到的均方误差(RMSE)都会增大;完全贝叶斯方法和部分贝叶斯方法在缺失比例较小时几乎相同,只在缺失比例为0.5时,前者明显优于后者.
Abstract:
The research explored the relative performance of Fully Bayesian method and Partially method in the estimation of growth curve model parameters. Only Simulation studies were used in the compassion in which four missing rates (0, 0.10, 0.30, and 0.50) were investigated. In each situation, 50 matrixes with missing response were generated and the index RMSE (root mean square error) were compared the two approaches. The results showed that: (1) the accuracy of parameter estimations of the two approaches were both affected by the missing rate, and as the increasing of missing rate, the bigger of RMSE. (2)When the missing rate is small, the RMSEs of the two approaches were almost same, however, Fully Bayesian method got better than Partially method when missing rate came to 0.50.

参考文献/References:

[1] 刘红云, 孟庆茂. 纵向数据分析方法 [J]. 心理科学进展, 2003, 11(5): 586-92.
[2] 利特尔, 鲁宾. 缺失数据统计分析 [M]. 2版. 孙山泽, 译. 北京: 中国统计出版社, 2004.
[3] Nakai M, Ke W. Review of methods for handing missing data in longitudinal data analysis [J]. International Journal fo Mathematical Analysis, 2011, 5(1): 1-13.
[4] Graham J W. Missing data analysis: making it work in the real world [J]. The Annual Review of Psychology, 2009, 60: 549-76.
[5] Little R J A. Modeling the drop-out mechanism in repeated-measures studies [J]. Journal of the American Statistical Association, 1995, 90: 1112-1121.
[6] In Litter T D, Schnabel K U, Baumert J. Modeling longitudinal and multilevel data: practice issues, applied approaches and specific examples [C]. New Jersey: Laerence Erlbaum Associates, 2000: 15-32.
[7] Rubin D B. Multiple imputation for nonresponse in survers [M]. New York: Wiley, 1987.
[8] Ibrahim J G, Chen M H, Lipsitz S R, et al. Missing-data methods for gerneralized linear models: a comparative review [J]. Journal of American Statistical Association, 2005, 469(100): 332-346.
[9] Newman D A. Longitudinal modeling with randomly and systematically missing data: a simulation of Ad hoc, maximum likelihood, and multiple impution techniques [J]. Organizational Research Methods , 2003, 6(3): 328-362.
[10] Edwards W, Lindman H, Savage L J. Bayesian statistical inference for psychological research [J]. Psychological Review, 1963, 70: 193-242.
[11] Carrigan G, Barnett A G, et al. Compensating missing data from longitudinal studies using WinBUGS [J]. Journal of Statistical Software, 2007, 19(7): .
[12] Li Jinhui. Analysis of longitudinal data with missing values [D]. California : University of California, 2006.
[13] Zhang Zhiyong, Hamagami F, Wang L J, et al. Bayesian analysis of longitudinal data using growth curve models [J]. International Journal of Behavioral Development, 2007, 31 (4): 374-83.
[14] Geman S, Geman D. Stochastic relaxation, Gibbs distributions, and the Bayesian retoration of image [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1984, 6: 721-741.
[15] Spiegelhalter D J, Thomas A, Best NG, et al. WinBUGS Version 1.4 User Manual [EB/OL].
[2011-12-16]. http: //www. mrc-bsu. cam. ac. uk/bugs/.
[16] R Development Core Team. R: a language and environment for statistical computing[EB/OL].
[2012-12-16]. http: //www. R-project. org/.
[17] 漆书青, 戴海琦, 丁树良. 现代教育与心理测量学原理 [M]. 北京: 高等教育出版社, 2002.

更新日期/Last Update: 1900-01-01