From: shahryar.rahman@gmail.com   
      
   Hi,   
   I am working on a project that intends to investigate the   
   implementation of a modern heuristic  (e.g. simulated annealing,   
   genetic algorithms or local search) to search through a space of   
   polynomial transformations and assign selections for a linear   
   regression.   
      
   I have read that standard statistical methods for finding suitable   
   transformations of regressors use hill-climbing algorithms to search   
   for the correct transformations for linear modelling. I have found   
   that alot of times techniques such as stepwise regression have been   
   used to select a subset of regressors  using a greedy algorithm.   
      
   BUT when this technique is used on a more complex model these   
   algorithms would fail to reach a global optimum.   
      
   I would like to know if by adopting a heuristic technique it may be   
   possible to provide better results.   
      
   (Could anyone post any suggestions/possible reading material/anything   
   that has been done along the same lines)   
      
   Thankyou,   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)