作者tokyo291 (工口工口)
看板Statistics
標題[問題] 存活分析和迴歸分析的問題
時間Tue Jan 8 18:04:27 2013
1.If Y is linear dependent on X1 and X2 where Y, X1 and X2
are normally distributed with Var(Y)=9, Var(X1)=1, Var(X2)=4,
ρ(Y,X1)=0.6, ρ(Y,X2)=0.5, ρ(X1,X2)=0.4, and we consider the
regression of Y on X1 and X2, i.e. Y=α+β1*X1+β2*X2+ε,ε~N(0,σ^2)
(A) If E(Y)=E(X1)=E(X2)=0, find the values of α, β1 and β2.
(B) How much is R^2 (coefficient of determination of using X1 and X2
to determine Y ) in the above regression model?
這邊根據目前學到的都是需要樣本來做估計,可是這邊給了母體參數就
不知道該怎麼做了...有試過作多變量Normal dist的轉換 可是好像沒甚麼用
2. Life table
interval 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10
# of deaths 0 5 7 10 14 16 15 14 16 18
# of censored 0 0 1 0 1 1 0 0 1 0
S(t):cumulative proportion surviving
Plese estimate the values of S(5.5).
^ ^ ^
這裡我想請問可以把S(5.5)=(S(6)+S(5))/2 這要做來估計嗎?
--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 140.116.52.26
1F:→ yhliu:1. 樣本資料迴歸計算的公式改用群體參數而已. 01/10 10:04
2F:→ yhliu:2. 應該用算術平均數嗎? 01/10 10:06
3F:→ tokyo291:謝謝回答!感覺用算術平均比較合理 01/10 14:32