Talk:SPC: random sampling from a stationary Gaussian process

I'd just like to say how I think it's funny you can roll 2 dice and get a result of 2.5! :) Those are some magical dice! -- good point! I was wondering how you can take 2 numbers get an average of 5 and a sigma of 2.3
 * is there a way to do this sampling in excel? Say using the CDF function or ERF function?  Or z-scores?
 * Example 2: this is kind of a p-value but would be more appropriate with a t-test. Here you know the sample means and sample standard deviations.
 * In example 2 a: were A and B actually drawn from different distributions? If not, then what can you say about the accuracy of the t-test?
 * why not use inverse error functions ?
 * in excel there is Tools->Data Analysis -> Random Number Generation that will generate any kind and number of random numbers you like..