There are some assumptions for an OLS regression:

1. E(u_i ): The errors have zero mean

2. The variance of the errors is constant (non-random) and finite over all values of X1

3. The errors are linearly independent of one another

4. Disturbances are homoscedastic (White test checks this)

5. Disturbances are not autocorrelated

Did you check for this?

In another statistical program (Eviews) you can control for autocorrelation and homo/heteroscedasticy. If you control for it, and the "problem" is gone, you can use an OLS regression. (If not, you can use it too, but you should mention it

). This should also be possible in Stata, but I never tried it

Check this link to find out how you can test your data in Stata:

http://www.ats.ucla.edu/stat/stata/webbooks/reg/chapter2/statareg2.htm