Useful Distributions Chuan Shi Suppose we have a population X ∼ N(µ, σ 2 ), and x1 , x2 , · · ·, xn is a n size sample of the population. The mean and variance of the sample are given as: x̄ = n n 1� 1 � xi and s2 = (xi − x̄)2 n i=1 n − 1 i=1 (1) The following distributions constructed with x̄ and s2 (or s) are very useful in Statistics. They could be helpful for your homework. � σ2 x¯ ∼ N µ, n � or x̄ − µ σ ∼ N(0, 1) √ n x̄ − µ s ∼ t(n − 1) √ n (2) (3) (n − 1)s2 ∼ χ2 (n − 1) (4) 2 σ Suppose there are two populations X ∼ N(µ1 , σ12 ) and Y ∼ N(µ2 , σ22 ), and a n1 size sample of X, and a n2 size sample of Y . Means of the two samples are x̄ and ȳ, respectively. Variances of the two samples are s21 and s22 , respectively. Then, (x̄ − µ1 ) − (ȳ − µ2 ) � σ12 σ22 + n1 n2 (x̄ − µ1 ) − (ȳ − µ2 ) � s2 s2 + n1 n2 where s2 = ∼ N(0, 1) (5) ∼ t(n1 + n2 − 2) (6) (n1 − 1)s21 + (n2 − 1)s22 n1 + n2 − 2 The distribution (6) above can be used in the Hypothesis Test of means of two populations, in case that σ1 and σ2 are unknown, but can be assumed equal. � � s21 σ2 � 1 � ∼ F (n1 − 1, n2 − 1) s22 σ22 1 (7) MIT OpenCourseWare http://ocw.mit.edu 2.854 / 2.853 Introduction to Manufacturing Systems Fall 2010 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.