Đang chuẩn bị nút TẢI XUỐNG, xin hãy chờ
Tải xuống
Tham khảo tài liệu 'a course in mathematical statistics phần 6', ngoại ngữ, ngữ pháp tiếng anh phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | 262 11 Sufficiency and Related Theorems each one of the n different ways in which the t successes can occur. Then if there are values of 9 for which particular occurrences of the t successes can happen with higher probability than others we will say that knowledge of the positions where the t successes occurred is more informative about 9 than simply knowledge of the total number of successes t. If on the other hand all possible outcomes given the total number of successes t have the same probability of occurrence then clearly the positions where the t successes occurred are entirely irrelevant and the total number of successes t provides all possible information about 9. In the present case we have P9X Xi . Xn xn T t P9IXi Xi . Xn Xn I T t -i 9 11 n n Ì p9 t t Pẹ X1 x1 . Xn Xn Pg T t if X1 . . . Xn t and zero otherwise and this is equal to 9x1 1 -9 1-X1 . . . 9x 1 -9 1-x n Í . _ n-i n 9 1 - 9 V17 n-t 9 1 - 9 n . n-t n 9 1 - 9 V17 1 c n V17 if X1 xn t and zero otherwise. Thus we found that for all x1 . xnsuch that xj 0 or 1 j 1 . n and n y xj t p0 X1 x1 . j 1 Xn Xn I T t V 1 7 independent of 9 and therefore the total number of successes t alone provides all possible information about 9. This example motivates the following definition of a sufficient statistic. DEFINITION 1 LetXj j 1 . nbe i.i.d. r.v. s with p.d.f. 0 0 91 . 9 e Qc Rr and let T T1 . Tm where Tj Tj X1 . Xn j 1 . m are statistics. We say that T is an m-dimensional sufficient statistic for the family F 0 0 e Q or for the parameter 0 if the conditional distribution of X1 . Xn given T t is independent of 0 for all values of t actually for almost all a.a. t that is except perhaps for a set N in Rm of values of t such that P0 T e N 0 for all 0 e Q where Pg denotes the probability function associated with the p.d.f. 0 . REMARK 1 Thus T being a sufficient statistic for 9 implies that every measurable set A in nr P0 X1 . Xn e AIT t is independent of 0 for a.a. 11.1 Sufficiency Definition and Some Basic Results
