question 28 spring 1997

Donald Morrison ( donald_morrison@es.adp.com )
Wed, 14 Apr 1999 09:54:51 -0400


40 grouped losses

interval ($000)

(1,4/3) 16

(4/3,2) 10

[2,4) 10

[4,infinity) 4


null hypothesis (Ho) is that random variable X underlying the

observed losses in thousands has the density function f(x) =

1/xsquared where 1<x<infinity


compute the exact value of the kolmogrov-smirnov




when i did the problem i used the midpoints of the interval for max

and got the right answer is that just a coincidence or is that


the an acceptable solution

thanks