UMVUE Geometric Distribution
up vote
2
down vote
favorite
I am trying to find the UMVUE for the parameter $p$ for an n i.i.d geometric distribution:
$(1-p)^{x-1}p$ for $x=1,2,…$ and $0<p<1$
and found that:
$P(X_1=1)$ is an unbiased estimator , so let $w=I[X_1=1]$ be my unbiased estimator and since $sum_i X_i=t$ is complete and sufficient statistic for geometric distribution, I can improve my unbiased estimator as follows:
$E[wmidsum_i X_i=t] = P(X_1=1midsum_i X_i=t) = frac{P(X_1=1,sum_i X_i=t-1)}{P(sum_i X_i=t)}$
So I have two questions now:
what is the pdf for $sum_i X_i=t-1$ ? .. I know it is negative binomial but can't write it correctly
and my second question is what is the variance of this modified unbiased estimator and does it achieve the Cramer-Rao lower bound ?
self-learning statistical-inference
add a comment |
up vote
2
down vote
favorite
I am trying to find the UMVUE for the parameter $p$ for an n i.i.d geometric distribution:
$(1-p)^{x-1}p$ for $x=1,2,…$ and $0<p<1$
and found that:
$P(X_1=1)$ is an unbiased estimator , so let $w=I[X_1=1]$ be my unbiased estimator and since $sum_i X_i=t$ is complete and sufficient statistic for geometric distribution, I can improve my unbiased estimator as follows:
$E[wmidsum_i X_i=t] = P(X_1=1midsum_i X_i=t) = frac{P(X_1=1,sum_i X_i=t-1)}{P(sum_i X_i=t)}$
So I have two questions now:
what is the pdf for $sum_i X_i=t-1$ ? .. I know it is negative binomial but can't write it correctly
and my second question is what is the variance of this modified unbiased estimator and does it achieve the Cramer-Rao lower bound ?
self-learning statistical-inference
add a comment |
up vote
2
down vote
favorite
up vote
2
down vote
favorite
I am trying to find the UMVUE for the parameter $p$ for an n i.i.d geometric distribution:
$(1-p)^{x-1}p$ for $x=1,2,…$ and $0<p<1$
and found that:
$P(X_1=1)$ is an unbiased estimator , so let $w=I[X_1=1]$ be my unbiased estimator and since $sum_i X_i=t$ is complete and sufficient statistic for geometric distribution, I can improve my unbiased estimator as follows:
$E[wmidsum_i X_i=t] = P(X_1=1midsum_i X_i=t) = frac{P(X_1=1,sum_i X_i=t-1)}{P(sum_i X_i=t)}$
So I have two questions now:
what is the pdf for $sum_i X_i=t-1$ ? .. I know it is negative binomial but can't write it correctly
and my second question is what is the variance of this modified unbiased estimator and does it achieve the Cramer-Rao lower bound ?
self-learning statistical-inference
I am trying to find the UMVUE for the parameter $p$ for an n i.i.d geometric distribution:
$(1-p)^{x-1}p$ for $x=1,2,…$ and $0<p<1$
and found that:
$P(X_1=1)$ is an unbiased estimator , so let $w=I[X_1=1]$ be my unbiased estimator and since $sum_i X_i=t$ is complete and sufficient statistic for geometric distribution, I can improve my unbiased estimator as follows:
$E[wmidsum_i X_i=t] = P(X_1=1midsum_i X_i=t) = frac{P(X_1=1,sum_i X_i=t-1)}{P(sum_i X_i=t)}$
So I have two questions now:
what is the pdf for $sum_i X_i=t-1$ ? .. I know it is negative binomial but can't write it correctly
and my second question is what is the variance of this modified unbiased estimator and does it achieve the Cramer-Rao lower bound ?
self-learning statistical-inference
self-learning statistical-inference
edited Dec 1 '16 at 1:46
Momo
12k21430
12k21430
asked Dec 1 '16 at 1:10
Bassem
478
478
add a comment |
add a comment |
1 Answer
1
active
oldest
votes
up vote
0
down vote
For the first question only:
$P(X_1=1)=p$
$P(sum_{i=2}^{n}X_i=t-1)={t-2choose n-2}p^{n-1}(1-p)^{t-n}$, $t=n,n+1...$
$P(sum_{i=1}^{n}X_i=t)={t-1choose n-1}p^{n}(1-p)^{t-n}$, $t=n,n+1...$
So the UMVUE is $hat p=frac{n-1}{sum_{i=1}^{n} X_i-1}$
For CRLB you may look here.
But for the variance of the UMVUE:
$Var(hat p)=sum_{t=n}^infty left(frac{n-1}{t-1}-pright)^2 {t-1choose n-1}p^n(1-p)^{t-n}$
I'm afraid I was not able to get a closed form. Neither it worked for $E(hat p^2)$
Maybe somebody else can step in.
Can we find a closed form for the variance or E(p^2) when n=2 ? that is $ hat p=frac{n-1}{sum_{i=1}^{2} X_i-1}$ and how did you develop the variance summation? thanks
– Bassem
Dec 1 '16 at 12:24
$Y=sum_{i=1}^n X_i$ is Negative Binomial, $T=frac{n-1}{Y-1}$ is unbiased ($E[T]=p$), so $Var(T)=E[(T-E[T])^2]=Eleft[left(frac{n-1}{Y-1}-pright)^2right]=sum_{t=n}^inftyleft(frac{n-1}{t-1}-pright)^2 P(Y=t)$
– Momo
Dec 1 '16 at 14:33
Also, for $n=2$ you have $E[hat{p}^2]=frac{p^2log(1/p)}{1-p}$ so $Var(hat p)=E[hat{p}^2]-p^2$
– Momo
Dec 1 '16 at 14:39
Great .. thanks for the help .. but could you show what form did you use to find $E[hat{p}^2]$ ?
– Bassem
Dec 1 '16 at 23:37
$E(hat p^2)=sum_{t=n}^infty left(frac{n-1}{t-1}right)^2 {t-1choose n-1}p^n(1-p)^{t-n}$ So for $n=2$ $E(hat p^2)=sum_{t=2}^infty frac{1}{t-1} p^2(1-p)^{t-2}=frac{p^2}{1-p}sum_{t=2}^infty frac{1}{t-1} (1-p)^{t-1}=frac{p^2}{1-p}sum_{i=1}^infty frac{1}{i} (1-p)^i$ The last series needs $sum_{i=1}^infty frac{x^i}{i}$, which is obtained by integrating $sum_{i=1}^infty u^{i-1}=frac{1}{1-u}$ term by term from $u=0$ to $x$ You might consider upvoting and accepting the answer, if it was useful for you.
– Momo
Dec 1 '16 at 23:50
|
show 1 more comment
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
0
down vote
For the first question only:
$P(X_1=1)=p$
$P(sum_{i=2}^{n}X_i=t-1)={t-2choose n-2}p^{n-1}(1-p)^{t-n}$, $t=n,n+1...$
$P(sum_{i=1}^{n}X_i=t)={t-1choose n-1}p^{n}(1-p)^{t-n}$, $t=n,n+1...$
So the UMVUE is $hat p=frac{n-1}{sum_{i=1}^{n} X_i-1}$
For CRLB you may look here.
But for the variance of the UMVUE:
$Var(hat p)=sum_{t=n}^infty left(frac{n-1}{t-1}-pright)^2 {t-1choose n-1}p^n(1-p)^{t-n}$
I'm afraid I was not able to get a closed form. Neither it worked for $E(hat p^2)$
Maybe somebody else can step in.
Can we find a closed form for the variance or E(p^2) when n=2 ? that is $ hat p=frac{n-1}{sum_{i=1}^{2} X_i-1}$ and how did you develop the variance summation? thanks
– Bassem
Dec 1 '16 at 12:24
$Y=sum_{i=1}^n X_i$ is Negative Binomial, $T=frac{n-1}{Y-1}$ is unbiased ($E[T]=p$), so $Var(T)=E[(T-E[T])^2]=Eleft[left(frac{n-1}{Y-1}-pright)^2right]=sum_{t=n}^inftyleft(frac{n-1}{t-1}-pright)^2 P(Y=t)$
– Momo
Dec 1 '16 at 14:33
Also, for $n=2$ you have $E[hat{p}^2]=frac{p^2log(1/p)}{1-p}$ so $Var(hat p)=E[hat{p}^2]-p^2$
– Momo
Dec 1 '16 at 14:39
Great .. thanks for the help .. but could you show what form did you use to find $E[hat{p}^2]$ ?
– Bassem
Dec 1 '16 at 23:37
$E(hat p^2)=sum_{t=n}^infty left(frac{n-1}{t-1}right)^2 {t-1choose n-1}p^n(1-p)^{t-n}$ So for $n=2$ $E(hat p^2)=sum_{t=2}^infty frac{1}{t-1} p^2(1-p)^{t-2}=frac{p^2}{1-p}sum_{t=2}^infty frac{1}{t-1} (1-p)^{t-1}=frac{p^2}{1-p}sum_{i=1}^infty frac{1}{i} (1-p)^i$ The last series needs $sum_{i=1}^infty frac{x^i}{i}$, which is obtained by integrating $sum_{i=1}^infty u^{i-1}=frac{1}{1-u}$ term by term from $u=0$ to $x$ You might consider upvoting and accepting the answer, if it was useful for you.
– Momo
Dec 1 '16 at 23:50
|
show 1 more comment
up vote
0
down vote
For the first question only:
$P(X_1=1)=p$
$P(sum_{i=2}^{n}X_i=t-1)={t-2choose n-2}p^{n-1}(1-p)^{t-n}$, $t=n,n+1...$
$P(sum_{i=1}^{n}X_i=t)={t-1choose n-1}p^{n}(1-p)^{t-n}$, $t=n,n+1...$
So the UMVUE is $hat p=frac{n-1}{sum_{i=1}^{n} X_i-1}$
For CRLB you may look here.
But for the variance of the UMVUE:
$Var(hat p)=sum_{t=n}^infty left(frac{n-1}{t-1}-pright)^2 {t-1choose n-1}p^n(1-p)^{t-n}$
I'm afraid I was not able to get a closed form. Neither it worked for $E(hat p^2)$
Maybe somebody else can step in.
Can we find a closed form for the variance or E(p^2) when n=2 ? that is $ hat p=frac{n-1}{sum_{i=1}^{2} X_i-1}$ and how did you develop the variance summation? thanks
– Bassem
Dec 1 '16 at 12:24
$Y=sum_{i=1}^n X_i$ is Negative Binomial, $T=frac{n-1}{Y-1}$ is unbiased ($E[T]=p$), so $Var(T)=E[(T-E[T])^2]=Eleft[left(frac{n-1}{Y-1}-pright)^2right]=sum_{t=n}^inftyleft(frac{n-1}{t-1}-pright)^2 P(Y=t)$
– Momo
Dec 1 '16 at 14:33
Also, for $n=2$ you have $E[hat{p}^2]=frac{p^2log(1/p)}{1-p}$ so $Var(hat p)=E[hat{p}^2]-p^2$
– Momo
Dec 1 '16 at 14:39
Great .. thanks for the help .. but could you show what form did you use to find $E[hat{p}^2]$ ?
– Bassem
Dec 1 '16 at 23:37
$E(hat p^2)=sum_{t=n}^infty left(frac{n-1}{t-1}right)^2 {t-1choose n-1}p^n(1-p)^{t-n}$ So for $n=2$ $E(hat p^2)=sum_{t=2}^infty frac{1}{t-1} p^2(1-p)^{t-2}=frac{p^2}{1-p}sum_{t=2}^infty frac{1}{t-1} (1-p)^{t-1}=frac{p^2}{1-p}sum_{i=1}^infty frac{1}{i} (1-p)^i$ The last series needs $sum_{i=1}^infty frac{x^i}{i}$, which is obtained by integrating $sum_{i=1}^infty u^{i-1}=frac{1}{1-u}$ term by term from $u=0$ to $x$ You might consider upvoting and accepting the answer, if it was useful for you.
– Momo
Dec 1 '16 at 23:50
|
show 1 more comment
up vote
0
down vote
up vote
0
down vote
For the first question only:
$P(X_1=1)=p$
$P(sum_{i=2}^{n}X_i=t-1)={t-2choose n-2}p^{n-1}(1-p)^{t-n}$, $t=n,n+1...$
$P(sum_{i=1}^{n}X_i=t)={t-1choose n-1}p^{n}(1-p)^{t-n}$, $t=n,n+1...$
So the UMVUE is $hat p=frac{n-1}{sum_{i=1}^{n} X_i-1}$
For CRLB you may look here.
But for the variance of the UMVUE:
$Var(hat p)=sum_{t=n}^infty left(frac{n-1}{t-1}-pright)^2 {t-1choose n-1}p^n(1-p)^{t-n}$
I'm afraid I was not able to get a closed form. Neither it worked for $E(hat p^2)$
Maybe somebody else can step in.
For the first question only:
$P(X_1=1)=p$
$P(sum_{i=2}^{n}X_i=t-1)={t-2choose n-2}p^{n-1}(1-p)^{t-n}$, $t=n,n+1...$
$P(sum_{i=1}^{n}X_i=t)={t-1choose n-1}p^{n}(1-p)^{t-n}$, $t=n,n+1...$
So the UMVUE is $hat p=frac{n-1}{sum_{i=1}^{n} X_i-1}$
For CRLB you may look here.
But for the variance of the UMVUE:
$Var(hat p)=sum_{t=n}^infty left(frac{n-1}{t-1}-pright)^2 {t-1choose n-1}p^n(1-p)^{t-n}$
I'm afraid I was not able to get a closed form. Neither it worked for $E(hat p^2)$
Maybe somebody else can step in.
edited Apr 13 '17 at 12:21
Community♦
1
1
answered Dec 1 '16 at 1:33
Momo
12k21430
12k21430
Can we find a closed form for the variance or E(p^2) when n=2 ? that is $ hat p=frac{n-1}{sum_{i=1}^{2} X_i-1}$ and how did you develop the variance summation? thanks
– Bassem
Dec 1 '16 at 12:24
$Y=sum_{i=1}^n X_i$ is Negative Binomial, $T=frac{n-1}{Y-1}$ is unbiased ($E[T]=p$), so $Var(T)=E[(T-E[T])^2]=Eleft[left(frac{n-1}{Y-1}-pright)^2right]=sum_{t=n}^inftyleft(frac{n-1}{t-1}-pright)^2 P(Y=t)$
– Momo
Dec 1 '16 at 14:33
Also, for $n=2$ you have $E[hat{p}^2]=frac{p^2log(1/p)}{1-p}$ so $Var(hat p)=E[hat{p}^2]-p^2$
– Momo
Dec 1 '16 at 14:39
Great .. thanks for the help .. but could you show what form did you use to find $E[hat{p}^2]$ ?
– Bassem
Dec 1 '16 at 23:37
$E(hat p^2)=sum_{t=n}^infty left(frac{n-1}{t-1}right)^2 {t-1choose n-1}p^n(1-p)^{t-n}$ So for $n=2$ $E(hat p^2)=sum_{t=2}^infty frac{1}{t-1} p^2(1-p)^{t-2}=frac{p^2}{1-p}sum_{t=2}^infty frac{1}{t-1} (1-p)^{t-1}=frac{p^2}{1-p}sum_{i=1}^infty frac{1}{i} (1-p)^i$ The last series needs $sum_{i=1}^infty frac{x^i}{i}$, which is obtained by integrating $sum_{i=1}^infty u^{i-1}=frac{1}{1-u}$ term by term from $u=0$ to $x$ You might consider upvoting and accepting the answer, if it was useful for you.
– Momo
Dec 1 '16 at 23:50
|
show 1 more comment
Can we find a closed form for the variance or E(p^2) when n=2 ? that is $ hat p=frac{n-1}{sum_{i=1}^{2} X_i-1}$ and how did you develop the variance summation? thanks
– Bassem
Dec 1 '16 at 12:24
$Y=sum_{i=1}^n X_i$ is Negative Binomial, $T=frac{n-1}{Y-1}$ is unbiased ($E[T]=p$), so $Var(T)=E[(T-E[T])^2]=Eleft[left(frac{n-1}{Y-1}-pright)^2right]=sum_{t=n}^inftyleft(frac{n-1}{t-1}-pright)^2 P(Y=t)$
– Momo
Dec 1 '16 at 14:33
Also, for $n=2$ you have $E[hat{p}^2]=frac{p^2log(1/p)}{1-p}$ so $Var(hat p)=E[hat{p}^2]-p^2$
– Momo
Dec 1 '16 at 14:39
Great .. thanks for the help .. but could you show what form did you use to find $E[hat{p}^2]$ ?
– Bassem
Dec 1 '16 at 23:37
$E(hat p^2)=sum_{t=n}^infty left(frac{n-1}{t-1}right)^2 {t-1choose n-1}p^n(1-p)^{t-n}$ So for $n=2$ $E(hat p^2)=sum_{t=2}^infty frac{1}{t-1} p^2(1-p)^{t-2}=frac{p^2}{1-p}sum_{t=2}^infty frac{1}{t-1} (1-p)^{t-1}=frac{p^2}{1-p}sum_{i=1}^infty frac{1}{i} (1-p)^i$ The last series needs $sum_{i=1}^infty frac{x^i}{i}$, which is obtained by integrating $sum_{i=1}^infty u^{i-1}=frac{1}{1-u}$ term by term from $u=0$ to $x$ You might consider upvoting and accepting the answer, if it was useful for you.
– Momo
Dec 1 '16 at 23:50
Can we find a closed form for the variance or E(p^2) when n=2 ? that is $ hat p=frac{n-1}{sum_{i=1}^{2} X_i-1}$ and how did you develop the variance summation? thanks
– Bassem
Dec 1 '16 at 12:24
Can we find a closed form for the variance or E(p^2) when n=2 ? that is $ hat p=frac{n-1}{sum_{i=1}^{2} X_i-1}$ and how did you develop the variance summation? thanks
– Bassem
Dec 1 '16 at 12:24
$Y=sum_{i=1}^n X_i$ is Negative Binomial, $T=frac{n-1}{Y-1}$ is unbiased ($E[T]=p$), so $Var(T)=E[(T-E[T])^2]=Eleft[left(frac{n-1}{Y-1}-pright)^2right]=sum_{t=n}^inftyleft(frac{n-1}{t-1}-pright)^2 P(Y=t)$
– Momo
Dec 1 '16 at 14:33
$Y=sum_{i=1}^n X_i$ is Negative Binomial, $T=frac{n-1}{Y-1}$ is unbiased ($E[T]=p$), so $Var(T)=E[(T-E[T])^2]=Eleft[left(frac{n-1}{Y-1}-pright)^2right]=sum_{t=n}^inftyleft(frac{n-1}{t-1}-pright)^2 P(Y=t)$
– Momo
Dec 1 '16 at 14:33
Also, for $n=2$ you have $E[hat{p}^2]=frac{p^2log(1/p)}{1-p}$ so $Var(hat p)=E[hat{p}^2]-p^2$
– Momo
Dec 1 '16 at 14:39
Also, for $n=2$ you have $E[hat{p}^2]=frac{p^2log(1/p)}{1-p}$ so $Var(hat p)=E[hat{p}^2]-p^2$
– Momo
Dec 1 '16 at 14:39
Great .. thanks for the help .. but could you show what form did you use to find $E[hat{p}^2]$ ?
– Bassem
Dec 1 '16 at 23:37
Great .. thanks for the help .. but could you show what form did you use to find $E[hat{p}^2]$ ?
– Bassem
Dec 1 '16 at 23:37
$E(hat p^2)=sum_{t=n}^infty left(frac{n-1}{t-1}right)^2 {t-1choose n-1}p^n(1-p)^{t-n}$ So for $n=2$ $E(hat p^2)=sum_{t=2}^infty frac{1}{t-1} p^2(1-p)^{t-2}=frac{p^2}{1-p}sum_{t=2}^infty frac{1}{t-1} (1-p)^{t-1}=frac{p^2}{1-p}sum_{i=1}^infty frac{1}{i} (1-p)^i$ The last series needs $sum_{i=1}^infty frac{x^i}{i}$, which is obtained by integrating $sum_{i=1}^infty u^{i-1}=frac{1}{1-u}$ term by term from $u=0$ to $x$ You might consider upvoting and accepting the answer, if it was useful for you.
– Momo
Dec 1 '16 at 23:50
$E(hat p^2)=sum_{t=n}^infty left(frac{n-1}{t-1}right)^2 {t-1choose n-1}p^n(1-p)^{t-n}$ So for $n=2$ $E(hat p^2)=sum_{t=2}^infty frac{1}{t-1} p^2(1-p)^{t-2}=frac{p^2}{1-p}sum_{t=2}^infty frac{1}{t-1} (1-p)^{t-1}=frac{p^2}{1-p}sum_{i=1}^infty frac{1}{i} (1-p)^i$ The last series needs $sum_{i=1}^infty frac{x^i}{i}$, which is obtained by integrating $sum_{i=1}^infty u^{i-1}=frac{1}{1-u}$ term by term from $u=0$ to $x$ You might consider upvoting and accepting the answer, if it was useful for you.
– Momo
Dec 1 '16 at 23:50
|
show 1 more comment
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2038222%2fumvue-geometric-distribution%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown