This topic begs a few questions, starting with "Why do you need it?"
What has the cost of a project got to do with it's price?
Isn't there some intrinsic value of the project to the client that is totally unrelated to the cost of delivering it?
And as far as training on estimates, that implies an interest in developing "accurate estimates" for the future. That phrase should be seen as an oxymoron and banished from the PM phrasebook. Learning comes from identifying sources of deviation from expectations, but no two circumstances are ever the same. Your projects can't and shouldn't depend on "better estimates" but better methods of dealing with the uncertainty that is inherent and unavoidable in all project work.
Gee -- now that I've lobbed that grenade into the discussion, I almost wish I wasn't on my way out to a three week vacation with limited access to the net and the responses it engenders.