[商管] 統計geometric分配MLE的期望值變異數

看板Grad-ProbAsk作者tokyo291 (工口工口)時間11年前 (2013/04/08 21:51)推噓3(3推 0噓 4→)

留言7則, 3人參與討論串1/1

For a geometric distribution, it is known that Pr(X=k)=p*(1-p)^(k-1) where k is the total number of trials required to reach the first success. p的MLE為Xbar的倒數要求MLE的期望值和變異數我想說先用負二項去找E(1/Y) Y是X1,...,Xn的總合可是還是卡住了QQ -- ※ 發信站: 批踢踢實業坊(ptt.cc) ◆ From: 218.164.77.41

推

goshfju

04/09 07:02, , 1^F

04/09 07:02, 1^F

→

goshfju

04/09 07:03, , 2^F

04/09 07:03, 2^F

→

tokyo291

04/09 14:36, , 3^F

04/09 14:36, 3^F