At the Desert Research Institute we have been forecasting indirect cost
recovery (IDC) for many years by a method that combines actual year to
date revenue, forecasted revenue on current projects, and forecasted
revenue on pending proposals.
For current projects we have developed a spreadsheet program that
divides the current IDC balance by the number of months remaining on a
project and spreads it over the remaining months.  For pending proposals
we use this same method but modify the algorithm to include the
probability of funding and to incorporate the estimated starting and
ending dates of the proposed project.  The probability of funding factor
is estimated based on our actual experience with a sponsor as well as
other factors.  The level of detail required and the manual QA analysis
of individual projects make the process somewhat time consuming but for
a small research institution as ours, it is not prohibitive.
This IDC forecasting model is useful for short-term (1 year or less) but
it's accuracy decreases as the time horizon extends significantly
outward due to the ending of current projects and lack of data for new,
currently undeveloped proposals.
Finally , we average the results of this modeling algorithm with the
extrapolated year-to-date actual data to increase the accuracy of the
forecast.
For long term (>1 year) we also analyze the historical trend data to
estimate IDC revenue.
If anyone has any questions or would like a copy of our spreadsheet
program please let me know.

--
John Bees  xxxxxx@dri.edu
Director of Research Support
Desert Research Institute
Phone 702-673-7381  Fax 702-673-7397
http://www.dri.edu:/Admin/research/rshome.html