There are many recent discussions about using p-value in the large sample size (i.e 10,000). Most of these discussions say that p-value give a significant results in the case of large sample size because its based on sample size. There are many recommendations like reporting confidence intervals with p-value, using effect sample size, using what called D-value,...etc. The question here in the case of real data like RNA-seq data analysis, we will have large sample size and we can't choose the sample size (i.e. results for human gene expression levels). What is the best way to check the significant in the statistical analysis, for example in multiple linear regression analysis, how can we test the significant of the estimated coefficients? The following article discussed using D-value instead of p-value because it doesn't rely on the sample size, so it will give accurate results than p-value. I wonder if anyone can help with using this measure in regression.
Question: D-value instead of P-value in large sample size
3.3 years ago by
M K • 490
M K • 490 wrote:
ADD COMMENT • link •
Please log in to add an answer.
Powered by Biostar version 2.3.0
Traffic: 944 users visited in the last hour