XPost: nyc.transit, nyc.politics, nyc.general 
 From: tls@panix.com 
  
 In article <854fb0pp49krt5pub10idhlg9vn01aho80@4ax.com>, 
 Obwon   wrote: 
 >On 27 May 2004 23:27:09 -0400, tls@panix.com (Thor 
 >Lancelot Simon) wrote: 
 > 
 >>In article <57ecb0pn2b0opq8r2064386tb7kidt6hv7@4ax.com>, 
 >>Obwon   wrote: 
 >>> 
 >>>  But that's because they can't look at the data where 
 >>>the owner of the car was summoned by the alarm and the 
 >>>theft was therefore prevented!   Or do they have that 
 >>>data too? 
 >> 
 >>How dumb are you?  Do you really think that insurers only insure 
 >>cars that are stolen? 
 > 
 >  And you're smart for thinking that only insured cars 
 >get stolen? 
  
 Given that only a trivial understanding of statistics and a quick 
 glance at the data are required, I'd say no.  No, you don't have to 
 be particularly smart to understand why it is reasonable to use 
 insurance data for this purpose. 
  
 On the other hand, you really do have to be droolingly, ponderously 
 stupid to think that knowing which cars in a given sample _were_ 
 stolen doesn't tell you which ones _weren't_. 
  
 -- 
  Thor Lancelot Simon                                       tls@rek.tjls.com 
    But as he knew no bad language, he had called him all the names of common 
  objects that he could think of, and had screamed: "You lamp!  You towel! 
 You 
  plate!" and so on.              --Sigmund Freud 
  
 --- SoupGate-Win32 v1.05 
  * Origin: you cannot sedate... all the things you hate (1:229/2)