Difference of two means significantly different to zero

[John Lei]

Hey guys, I need your help. So I calculated the mean of 2 groups and then calculated the difference of these two means in excel and know I want to know if this difference is significantly different than zero. But I had to do preserve and restore before calculating the second mean. So I did summarize to see the result of the two means but because of restore the variable are not in the dataset anymore. So for example I want to know if the difference of the two means, e.g. 0,4 ist sig. different to zero. How to I calculate this? And to I have to do this for every single value or is there a faster way? I hope you understand my problem.
Code:
preserve
calculation of the mean of a given variable (e.g. the return) of every stock for condition 1
calulation of first mean (the mean of all means)
summarize
restore (in order to not shorten the dataset)

preserve
calculation of the mean of a given variable (e.g. the return) of every stock for condition 2
calculation of second mean (the mean of all means)
summarize
restore

then because of restore the mean variable 1 and 2 do not exist anymore in the dataset. Now I want to know if the difference of mean 1 and 2 ist sig. different to the value zero. Another problem ist that the standard deviation of the means is zero because I calculated both means from the means of every stock in my dataset.

Thanks!

[Leonardo Guizzetti]

I'm not clear on exactly what you're doing to manipulate your dataset or why, but what's wrong with a two-group t-test?

[John Lei]

I am dropping observations that do not have the dummy==1 for example so I have to do the presverve restore stuff so that the original dataset is restored. So I gave every fund the dummy value of 1 if it occurs the first time (same as a tagvariable) and then I calculate the mean of the returns of every fund and drop it if it has the dummy value 0 so that every fund is only one time in my dataset. and then I calculate the mean of all means of the funds. So my standard deviation of the mean of means will be zero and with zero SD I can not calculate the ttest or p-value