• If the sample size is small(n˂30), the value of S2 fluctuate
considerably from sample to sample and the distribution of the random variable (X̅ -μ)/(S/√n) is no longer a standard normal distribution. We are now dealing with the distribution of statistic that we shall call T, where • T = (X̅-μ)/(S/√n) • The distribution of T is similar to the distribution of Z in that they both are symmetric about a mean of zero. Both distributions are bell-shaped, but the t distribution is more variable, owing to the fact that the T values depend on the on the fluctuations of two of two quantities X̅ and S2, whereas the Z values depend only on the changes of X̅ from sample to sample. The distribution of T differs from that of Z in that the variance of T depends on the sample size and always greater than 1. Example • A manufacture of light bulbs claims that his bulbs will burn on the average 500 hours. To maintain this average, he tests 25 bulbs each month. If the computed t values falls between –t0.05 and t0.05, he is satisfied with his claim. What conclusion should he draw from a sample that has mean x̅ =518 hours and standard deviation s = 40 hours? Assuming the distribution of burning times is approximately normal. • Solution • From table V we find t0.05 =1.711 for 24 degrees of freedom. Therefore, the manufacture is satisfied with claim if a sample of 25 bulbs yield a t value between -1.711 and 1.711. If μ = 500, then • t = (518 -500)/(40/√25) = 2.25 • a value well above 1.711. The probability of obtaining a t value, with ν =24 equal or greater than 2.25 is approximately 0.02. If μ ˃ 500, the value of t computed from the sample would be more reasonable. Hence the manufacture is likely to conclude that his bulbs are a better product than he thought. F Distribution
• One of the most important distribution in applied statistics is the F
Distribution. The statistic F is defined to be the ratio of two independent chi-square random variables, each divided by their degrees of freedom. Hence • F = (U/ν1)/(V/ν2) • where U and V are independent random variables having chi-square distributions with ν1 and ν2 degrees of freedom, respectively. • Let us define fα to be a particular value f of the random variable F above which we find an area equal to α. • Writing fα (ν1,ν2) for fα with ν1 and ν2 degrees of freedom we obtain • f1-α(ν1,ν2) = 1/ fα (ν2,ν1) • f0.95(6,10) = 1/ f0.05 (10,6)=1/4.06=0.246 • If S12 and S22 are the variances of independent random samples of size n1 and n2 taken from normal populations with variances σ12 and σ22 , respectively, then • F = (S12/ σ12)/( S22/ σ22) =( σ22S12)/( σ12 S22)
(Lecture Notes in Computer Science 6000 - Security and Cryptology) Ed Gerck (Auth.), David Chaum, Markus Jakobsson, Ronald L. Rivest, Peter Y. A. Ryan, Josh Benaloh, Miroslaw Kutylowski, Ben Adida (Ed