Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.


Luis A. Afonso
Posts:
4,743
From:
LIsbon (Portugal)
Registered:
2/16/05


Where the lines do cross each other (II)
Posted:
Apr 22, 2014 4:29 PM


Where the lines do cross each other (II)
At my Aprl 8, 6:53 thread a method was proposed to find out, among two different Distributions, and through a given parameter, the original source. I had shown, that this aim can be just achieved, with 95% Power, by using a crescent sample sizes list, that a singular point, were the H0 0.95 fractile descending Skewness values crosses the 0.05 fractile ascending H1, the common sample size can be promptly elected. To illustrate an important point where only occasionally the 5% monotonic behavior ends we run a routine regarding the Chisquared 5 degrees of freedom Skewness: _____Exp. ___a_________b__________c_____ _____20___ 0.036_______0.058____0.102__ _____25___ 0.067______ 0.039_____ 0.016__ _____30___ 0.164______ 0.113_____ 0.172__ _____35___ 0.224______ 0.208_____ 0.189__ _____40___ 0.267______ 0.252_____ 0.275__ _____45___ 0.292______ 0.292_____0.295__ _____50___ 0.349______0.327_____ 0.327__ ____ 55___ 0.346 (.369)_0.372_____0.393___ ____ 60___ 0.390______0.410_____0.397___ ____ 65___ 0.415______0.424_____0.421___ ____ 70___ 0.459______0.463_____ 0.441___ ____75___ 0.452(.475)_ 0.479_____0.489____ ____80___ 0.492______ 0.503_____0.483(.492) ____85___ 0.510______ 0.515_____0.494__ ____90___ 0.503(.524)_ 0.525_____0.530__ ____95___ 0.537______0.534_____0.538__ ___100___ 0.565______0.558_____0.567__
A simple linear interpolation is enough to reestablish the desired decreasing pattern.
Luis A. Afonso



