Rrerku

Lowest total scrabble score

What exactly color does ozone gas have?

Probability that THHT occurs in a sequence of 10 coin tosses

Add big quotation marks inside my colorbox

Limits and Infinite Integration by Parts

Does malloc reserve more space while allocating memory?

What does "Scientists rise up against statistical significance" mean? (Comment in Nature)

Yosemite Fire Rings - What to Expect?

How do apertures which seem too large to physically fit work?

Angel of Condemnation - Exile creature with second ability

What to do when eye contact makes your subordinate uncomfortable?

What if a revenant (monster) gains fire resistance?

Using substitution ciphers to generate new alphabets in a novel

putting logo on same line but after title, latex

Do we have to expect a queue for the shuttle from Watford Junction to Harry Potter Studio?

What is the highest possible scrabble score for placing a single tile

Why can Carol Danvers change her suit colours in the first place?

Can disgust be a key component of horror?

What is Cash Advance APR?

Quasinilpotent , non-compact operators

What if you are holding an Iron Flask with a demon inside and walk into Antimagic Field?

Multiplicative persistence

Plot of a tornado-shaped surface

System.QueryException unexpected token

Why is this estimator biased?

How do I check for bias of an estimator?Bootstrap confidence interval for a biased estimatorConceptual question about bias of an estimatorEstimator, unbiased or biasedCalculating risk for a density estimatorLinear model with biased estimatorBias Correction for Estimator with known biasWhy is OLS estimator of AR(1) coefficient biased?Cramer-Rao lower bound for biased estimatorVariance of distribution for maximum likelihood estimator

1

1

$begingroup$

$X_1,X_2,..,X_n$ are iid $sim Poisson(mu)$

than the MLE for $theta=e^-mu$ is $hat theta =e^-bar x$

Why is this considered to be biased for $theta$?

Is $E[hat theta]$ not $theta$ ?

as

$E[bar x]= mu$

estimation poisson-distribution bias

share|cite|improve this question

asked 1 hour ago

QualityQuality

29019

$endgroup$

• 
  
  
  

  
  3
  

  
  
  
  
  
  

  
  $begingroup$
  E[f(X)] != f(E[X]) in general.
  $endgroup$
  – The Laconic
  1 hour ago
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I know that general result but I am still confused how to do it here since the sum of poisson is also poisson with mean $n mu$
  $endgroup$
  – Quality
  1 hour ago
  

  
  
  
  

add a comment | 

1

1

$begingroup$

$X_1,X_2,..,X_n$ are iid $sim Poisson(mu)$

than the MLE for $theta=e^-mu$ is $hat theta =e^-bar x$

Why is this considered to be biased for $theta$?

Is $E[hat theta]$ not $theta$ ?

as

$E[bar x]= mu$

estimation poisson-distribution bias

share|cite|improve this question

asked 1 hour ago

QualityQuality

29019

$endgroup$

• 
  
  
  

  
  3
  

  
  
  
  
  
  

  
  $begingroup$
  E[f(X)] != f(E[X]) in general.
  $endgroup$
  – The Laconic
  1 hour ago
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I know that general result but I am still confused how to do it here since the sum of poisson is also poisson with mean $n mu$
  $endgroup$
  – Quality
  1 hour ago
  

  
  
  
  

add a comment | 

$begingroup$

$X_1,X_2,..,X_n$ are iid $sim Poisson(mu)$

than the MLE for $theta=e^-mu$ is $hat theta =e^-bar x$

Why is this considered to be biased for $theta$?

Is $E[hat theta]$ not $theta$ ?

as

$E[bar x]= mu$

estimation poisson-distribution bias

share|cite|improve this question

asked 1 hour ago

QualityQuality

29019

$endgroup$

share|cite|improve this question

asked 1 hour ago

QualityQuality

29019

share|cite|improve this question

asked 1 hour ago

QualityQuality

29019

asked 1 hour ago

QualityQuality

29019

asked 1 hour ago

QualityQuality

29019

2 Answers
2

active

oldest

votes

3

$begingroup$

Recall that the moment generating function of $X sim operatornamePoisson(mu)$ is
$$
E[e^s X] = exp(mu(e^s - 1))
$$
for all $s in mathbbR$

Proof.
We just compute:
$$
beginaligned
E[e^s X]
&= sum_k=0^infty e^s k e^-mu fracmu^kk! \
&= e^-mu sum_k=0^infty fracleft(mu e^sright)^kk! \
&= e^-mu e^mu e^s \
&= exp(mu(e^s - 1)).
endaligned
$$
We will use this result with $s = -1/n$.

If $X_1, ldots, X_n sim operatornamePoisson(mu)$ are i.i.d. and
$$
widehattheta
= expleft(-frac1n sum_i=1^n X_iright)
= prod_i=1^n expleft(-fracX_inright),
$$
then, using independence,
$$
beginaligned
E[widehattheta]
&= prod_i=1^n Eleft[e^-X_i / nright] \
&= prod_i=1^n exp(mu(e^-1/n - 1)) \
&= exp(n mu(e^-1/n - 1)) \
&neq e^-mu,
endaligned
$$
so $widehattheta$ is not an unbiased estimator of $e^-mu$

share|cite|improve this answer

answered 1 hour ago

Artem MavrinArtem Mavrin

1,066710

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Thank you, makes perfect sense
  $endgroup$
  – Quality
  1 hour ago
  

  
  
  
  

add a comment | 

3

$begingroup$

As an aside from finding the exact expectation, you can use Jensen's inequality, which says that for a random variable $X$ and a convex function $g$,

$$Eleft[g(X)right]ge gleft(Eleft[Xright]right)$$

, provided the expectations exist.

Verify that $g(x)=e^-x$ is a convex function from the fact that $g''>0$.

The equality does not hold because $g$ is not an affine function or a constant function.

share|cite|improve this answer

answered 1 hour ago

StubbornAtomStubbornAtom

2,7771532

$endgroup$

add a comment | 

Your Answer

StackExchange.ifUsing("editor", function ()
return StackExchange.using("mathjaxEditing", function ()
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix)
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
);
);
, "mathjax-editing");

StackExchange.ready(function()
var channelOptions =
tags: "".split(" "),
id: "65"
;
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function()
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled)
StackExchange.using("snippets", function()
createEditor();
);

else
createEditor();

);

function createEditor()
StackExchange.prepareEditor(
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
bindNavPrevention: true,
postfix: "",
imageUploader:
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
,
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
);



);

draft saved

draft discarded

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

Name

Email

Required, but never shown

StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstats.stackexchange.com%2fquestions%2f398954%2fwhy-is-this-estimator-biased%23new-answer', 'question_page');

);

Post as a guest

Name

Email

Required, but never shown

3

$begingroup$

Recall that the moment generating function of $X sim operatornamePoisson(mu)$ is
$$
E[e^s X] = exp(mu(e^s - 1))
$$
for all $s in mathbbR$

Proof.
We just compute:
$$
beginaligned
E[e^s X]
&= sum_k=0^infty e^s k e^-mu fracmu^kk! \
&= e^-mu sum_k=0^infty fracleft(mu e^sright)^kk! \
&= e^-mu e^mu e^s \
&= exp(mu(e^s - 1)).
endaligned
$$
We will use this result with $s = -1/n$.

If $X_1, ldots, X_n sim operatornamePoisson(mu)$ are i.i.d. and
$$
widehattheta
= expleft(-frac1n sum_i=1^n X_iright)
= prod_i=1^n expleft(-fracX_inright),
$$
then, using independence,
$$
beginaligned
E[widehattheta]
&= prod_i=1^n Eleft[e^-X_i / nright] \
&= prod_i=1^n exp(mu(e^-1/n - 1)) \
&= exp(n mu(e^-1/n - 1)) \
&neq e^-mu,
endaligned
$$
so $widehattheta$ is not an unbiased estimator of $e^-mu$

share|cite|improve this answer

answered 1 hour ago

Artem MavrinArtem Mavrin

1,066710

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Thank you, makes perfect sense
  $endgroup$
  – Quality
  1 hour ago
  

  
  
  
  

add a comment | 

3

$begingroup$

Recall that the moment generating function of $X sim operatornamePoisson(mu)$ is
$$
E[e^s X] = exp(mu(e^s - 1))
$$
for all $s in mathbbR$

Proof.
We just compute:
$$
beginaligned
E[e^s X]
&= sum_k=0^infty e^s k e^-mu fracmu^kk! \
&= e^-mu sum_k=0^infty fracleft(mu e^sright)^kk! \
&= e^-mu e^mu e^s \
&= exp(mu(e^s - 1)).
endaligned
$$
We will use this result with $s = -1/n$.

If $X_1, ldots, X_n sim operatornamePoisson(mu)$ are i.i.d. and
$$
widehattheta
= expleft(-frac1n sum_i=1^n X_iright)
= prod_i=1^n expleft(-fracX_inright),
$$
then, using independence,
$$
beginaligned
E[widehattheta]
&= prod_i=1^n Eleft[e^-X_i / nright] \
&= prod_i=1^n exp(mu(e^-1/n - 1)) \
&= exp(n mu(e^-1/n - 1)) \
&neq e^-mu,
endaligned
$$
so $widehattheta$ is not an unbiased estimator of $e^-mu$

share|cite|improve this answer

answered 1 hour ago

Artem MavrinArtem Mavrin

1,066710

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Thank you, makes perfect sense
  $endgroup$
  – Quality
  1 hour ago
  

  
  
  
  

add a comment | 

$begingroup$

Recall that the moment generating function of $X sim operatornamePoisson(mu)$ is
$$
E[e^s X] = exp(mu(e^s - 1))
$$
for all $s in mathbbR$

Proof.
We just compute:
$$
beginaligned
E[e^s X]
&= sum_k=0^infty e^s k e^-mu fracmu^kk! \
&= e^-mu sum_k=0^infty fracleft(mu e^sright)^kk! \
&= e^-mu e^mu e^s \
&= exp(mu(e^s - 1)).
endaligned
$$
We will use this result with $s = -1/n$.

If $X_1, ldots, X_n sim operatornamePoisson(mu)$ are i.i.d. and
$$
widehattheta
= expleft(-frac1n sum_i=1^n X_iright)
= prod_i=1^n expleft(-fracX_inright),
$$
then, using independence,
$$
beginaligned
E[widehattheta]
&= prod_i=1^n Eleft[e^-X_i / nright] \
&= prod_i=1^n exp(mu(e^-1/n - 1)) \
&= exp(n mu(e^-1/n - 1)) \
&neq e^-mu,
endaligned
$$
so $widehattheta$ is not an unbiased estimator of $e^-mu$

share|cite|improve this answer

answered 1 hour ago

Artem MavrinArtem Mavrin

1,066710

$endgroup$

share|cite|improve this answer

answered 1 hour ago

Artem MavrinArtem Mavrin

1,066710

answered 1 hour ago

Artem MavrinArtem Mavrin

1,066710

answered 1 hour ago

Artem MavrinArtem Mavrin

1,066710

3

$begingroup$

As an aside from finding the exact expectation, you can use Jensen's inequality, which says that for a random variable $X$ and a convex function $g$,

$$Eleft[g(X)right]ge gleft(Eleft[Xright]right)$$

, provided the expectations exist.

Verify that $g(x)=e^-x$ is a convex function from the fact that $g''>0$.

The equality does not hold because $g$ is not an affine function or a constant function.

share|cite|improve this answer

answered 1 hour ago

StubbornAtomStubbornAtom

2,7771532

$endgroup$

add a comment | 

3

$begingroup$

As an aside from finding the exact expectation, you can use Jensen's inequality, which says that for a random variable $X$ and a convex function $g$,

$$Eleft[g(X)right]ge gleft(Eleft[Xright]right)$$

, provided the expectations exist.

Verify that $g(x)=e^-x$ is a convex function from the fact that $g''>0$.

The equality does not hold because $g$ is not an affine function or a constant function.

share|cite|improve this answer

answered 1 hour ago

StubbornAtomStubbornAtom

2,7771532

$endgroup$

add a comment | 

$begingroup$

As an aside from finding the exact expectation, you can use Jensen's inequality, which says that for a random variable $X$ and a convex function $g$,

$$Eleft[g(X)right]ge gleft(Eleft[Xright]right)$$

, provided the expectations exist.

Verify that $g(x)=e^-x$ is a convex function from the fact that $g''>0$.

The equality does not hold because $g$ is not an affine function or a constant function.

share|cite|improve this answer

answered 1 hour ago

StubbornAtomStubbornAtom

2,7771532

$endgroup$

share|cite|improve this answer

answered 1 hour ago

StubbornAtomStubbornAtom

2,7771532

answered 1 hour ago

StubbornAtomStubbornAtom

2,7771532

answered 1 hour ago

StubbornAtomStubbornAtom

2,7771532