In the old days this would be viewed as concocting data, either by
interpolation or some general smoothing device. It was somewhat
dishonest because one could not quantify the extra information
introduced by the creation. Today, there is another option available
and that is imputation. Stata has a few options for doing this.
What you need is constrained imputation, for example if you
observe the quarterly sum of the monthly observations, then the
constraint is that the monthly observations need to satisfy the
sum constraint. If this makes sense, then the answer, as far
as I know, is no.
