[R] R-dvel [robustness Simulation study of 2 sample test on several combination of factors ]

[Michael Friendly]

On 4/4/2016 9:15 PM, tan sj wrote:
> hi, i am new in this field.
>
>
> do
> favorite<http://stackoverflow.com/questions/36404707/simulation-study-of-2-sample-test-on-different-combination-of-factors#>
>
>
> If I wish to conduct a simulation on the robustness of two sample test by using R language, is that any ways in writing the code?
> There are several factors
> (sample sizes-(10,10),(10,25),(25,25),(25,50),(25,100),50,25),(50,100), (100,25),(100,100))
>
> (standard deviation ratio- (1.00, 1.50, 2.00, 2.50, 3.00 and 3.50))
> distribution of gamma distribution with unequal skewness and equal skewness
>
> I wish to test the pooled variance t test and welch t test and mann whitney by using the above combination of factors. But how can I combine them by using for loop or apply function??
> I am intending to use apply function but i am stucking. If i use for loop function, can i use for loop with vectors ?
> for (a in c(25,50,100)) #first group of sample sizes
> { for (b in c(25,50,100)) #second group of sample sizes
> { for (d in c(4,4.4,5,6,8)) #different SDs of first sample
> the above code is an example that I would like to modified but I found I have different sets of sample sizes.
>

Don't try to code this yourself-- you'll run into a mess.  Instead, use 
the lovely SimDesign package, https://github.com/philchalmers/SimDesign
designed for just this purpose.

There are also some nice tutorial examples on a wiki,
https://github.com/philchalmers/SimDesign/wiki

best,
-Michael