I have applied 3 methods say Method-1, Method-2 and Method-3 to a dataset.
I applied modified versions of these methods say Method-1M, Method-2M and Method-3M.
while I can calculate the error between
-Method-1 and Method-1M
- Method-2 and Method-2M
- Method-3 and Method-3M
to see if the modified versions are better or not, I can also have other several combinations for error check.
My question is, is there any statistical method (say ANOVA) that can tell me in a single go that a method is better in terms of low error as compared to existing methods as well as among the modified versions?
Not sure I understand, but I think cross-validation using your empirical is what you are after, or if you want to more generally say "here are the conditions when this approach preforms differently than this one" simulation with those conditions.
The question makes no sense as you put it.
You must specify, first of all, the dataset you have built (or used), what is your purpose and what are these "methods" you would like to use.
Dear. Aftab Khan
The following paper may be helpful for you as I applied three Models in the paper. The values of R2, AIC, and BIC decide which model is the best model.
Article Intellectual capital and firm performance: evidence from cer...
Good luck
Best regards
Dr. Muhammad Yousaf
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS