The estimation of interval for overall mean effect in Meta-analysis
Qin Chao-ying · Basic Sciences Journal of Textile Universities · 2013
In the random effect of Meta-analysis,by means of improving the distributed inaccuracy of the corresponding statistics in the t-distribution approach to get a more accurate confidence interval for the overall mean effect.Compared with the D-L approach,coverage probability and accuracy of the confidence interval calculated by the improved t-distribution approach are both increased;compared with the t-distribution approach,this new method has an approximated coverage probability but improved accuracy.So the method proposed in this paper is based on the t-distribution approach,and it′s simple and effective.