# Differencing two means, one with and one without SD?

If we have access to a mean value (without access to the actual data points) without a standard deviation, and access to another mean value (with the same limitation regarding data points) with a standard deviation, is it possible to obtain a difference mean between them that includes a standard deviation? If so, how?

Example:

Mean of Random X: 150 ± 2.56

Mean of Random Y: 125 ± ?

Mean of Y - Mean of X = ? ± ?

-
I don't think this question is appropriate for MO; see the faq (link on top of the page). In any case, the answer is no, to the extent that the question is meaningful. –  Harald Hanche-Olsen Jun 30 '10 at 16:26
Fair enough, I'll flag it and hopefully it will get closed D: Thanks for the answer though –  Jordan Jun 30 '10 at 17:29
This question would bear some clarification. You refer to "data points". Does that mean the means and standard deviations you refer to are based on a random sample from each of two populations? –  Michael Hardy Jun 30 '10 at 23:14