Joint PMF

2009-10-07 12:40 am
1 The random variables X and Y are described by a joint PDF of the form
fX,Y (x,y) = ce^(-8x^2-6xy-18y^2)
:
Find the means, variances, and the correlation coefficient of X and Y . Also,find the
value of the constant c.
更新1:

i dun really understand wt _ is ? sorry about it...can u explain a bit more...thanks a lot

回答 (2)

2009-10-07 1:23 am
✔ 最佳答案
1 We recognize the pdf as a bivariate normal distribution
f(x,y)=ce^(-q(x,y)) where
c=1/[2πσ_xσ_y√(1-ρ)],
q(x,y)=[x^2/σ_x^2-2ρxy/σ_xσ_y+y^2/σ_y^2]/2(1-ρ^2)
Compare, 1/[2(1-ρ^2)σ_x^2]=8,2ρ/[2(1-ρ^2)σ_xσ_y]=-6,1/[σ_y^2 2(1-ρ^2)]=18
So (1-ρ^2)σ_x^2=1/16,(1-ρ^2)σ_y^2=1/36,(1-ρ^2)σ_xσ_y=-ρ/6
(1-ρ^2)σ_xσ_y=1/24=>ρ=-1/4
σ_x^2=1/15 and σ_y^2=4/135
So c=1/[2πσ_xσ_y√(1-ρ)]=√135/π

2009-10-06 17:24:53 補充:
The means of X and Y are 0
2009-10-08 4:58 am
好明顯你同下面兩條賊頭係同學:
http://hk.knowledge.yahoo.com/my/my?show=HA05715238
http://hk.knowledge.yahoo.com/my/my?show=HA01090095
真有趣你們為了毀屍滅跡,同樣開了散戶得了答案後立即刪除問題,不理那個答案到底定錯。
你之後兩條問題我已經分別比左上面兩條賊頭,你可以問佢地對,但我唔保証。

2009-10-07 20:59:14 補充:
本來我都唔志在分數,可以大方再貼多答案一次,但係我實在討厭你地的零回應不理對錯的態度,所以都係慳返啦。如果心夠黑的話,我會一路唔出聲,然後等你再問新問題個時就分別派相同錯誤的答案比你等你老師捉。不過哩樣實在唔係我風格。所以我決定啦:
你以後問新問題時我一定會答,但只會答一次,而且一定有錯,會24小時內刪除。
開其他散戶再發問唔比我認出?哈哈!我已然認出這位老師的問題風格,除非你係香港同台灣yahoo消失,唔係我一定認得你。


收錄日期: 2021-04-26 13:45:02
原文連結 [永久失效]:
https://hk.answers.yahoo.com/question/index?qid=20091006000051KK00769

檢視 Wayback Machine 備份