I have been testing the association of particular gene sets with a disease (using the tool MAGMA Article MAGMA: generalized gene-set analysis of GWAS data
). In the output I get:- BETA: the regression coefficient of the variable
- BETA_STD: the semi-standardized regression coefficient, corresponding to the predicted change in Z-value given a change of one standard deviation in the predictor gene set / gene covariate (ie. BETA divided by the variable’s standard deviation)
- SE: the standard error of the regression coefficient
- P: p-value for the parameter / variable
I have some significant results and I now want to know how strong of an effect those results have. I believe that Beta_STD is a standardized regression coefficient and therefore an effect size measure; however, I do not know what a good or bad value is for Beta_STD. My results with the lowest p-values also have the highest Beta_STD (~0.03) so yea.
Thank you in advance!
Babak Jamshidi
Standardization is a good idea to compare. However, P-value of the model works always.
0 votes
0 thanks
David Eugene Booth
The attached book information may be of interest for your work.. David Booth
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression
More Jon Klono's questions See All
Similar questions and discussions