- 21 -                                         
               it is methodologically appropriate to use the                           
               comprehensive sample I collected to estimate                            
               marketability discounts even though approximately half                  
               the sample consists of post-valuation date                              
               announcements because there is no reason to believe                     
               that the underlying economics of private placement                      
               would have changed after the valuation date.                            
               We agree.  Further, petitioners have not alleged that the               
          observed lack of marketability discounts in similar private                  
          placement transactions would have changed after the valuation                
          date.                                                                        
               Additionally, after performing an identical statistical                 
          analysis on only the prevaluation portion of his data sample,                
          Dr. Bajaj reported a predicted discount similar to the discount              
          he predicted using his complete data.10  For the foregoing                   
          reasons, petitioners' motion will be denied, and an appropriate              
          order will be issued.                                                        
          IV.  Valuation of the Gift                                                   
               1.  Tax Affecting G&J's Earnings                                        
                    A.  Introduction                                                   
               The decision whether to tax affect G&J’s projected earnings             
          under the discounted cash-flow approach accounts for the most                
          significant differences between the parties' expert witnesses.               
          In fact, Dr. Bajaj repeated his analysis, substituting a                     
          10   In fact, the predicted discount associated with the                     
          prevaluation data was slightly lower than with the full sample.              
Page:  Previous   11   12   13   14   15   16   17   18   19   20   21   22   23   24   25   26   27   28   29   30   NextLast modified: May 25, 2011