From: nrajkiran@gmail.com   
      
   Hi,   
   I am post graduate student doing a project in the area of soft   
   computing.Th problem i am trying to solve  is that of predicting   
   operaional risk. I have a small dataset of 25 data points that includes   
      
   five input variables - System downtime, Number of employees, Data   
   Quality, Number of transactions, Number of losses, and one output   
   variable - the loss amount. Because i have a very small dataset, i went   
      
   through  the following process to generate additional data points:   
    step 1: Select one variable at a time and fit various distributions   
   over it.   
    step 2: Based upon the goodness of fit tests, select the best   
   distribution for each variable seperately.   
    step 3: Generate random numbers for each variable over the selected   
   distribution seperately.   
    step 4: Tabulate the values.   
   My question is how do we ensure the correleation among the variables   
   that was there in the original sample data over the randomely selected   
   data as well. Because the random numbers were generated seperately for   
   each variable i could not find any correlation among the variables that   
      
   was present in the original sample data.   
   please do give me some suggestions, i will be waiting for them.   
      
   regards,   
   Raj kiran   
      
   [ 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)