Question: D-value instead of P-value in large sample size
1
gravatar for M K
2.4 years ago by
M K460
United States
M K460 wrote:

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.

http://amstat.tandfonline.com/doi/abs/10.1080/00031305.2015.1069760

rna-seq R genome • 1.3k views
ADD COMMENTlink modified 2.4 years ago • written 2.4 years ago by M K460
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1205 users visited in the last hour