Scribd
Upload
71 views

General Concepts of Point Estimation

1. A point estimator is a single value that represents an estimate of an unknown parameter based on sample data. An unbiased estimator has an expected value equal to the true value of the parameter being estimated. 2. The maximum likelihood estimator is the value of the parameter(s) that maximizes the likelihood function, which describes the probability of obtaining the sample data given different possible parameter values. 3. Desirably, point estimators should be unbiased and have minimum variance; the maximum likelihood estimator often has these desirable properties, making it a commonly used method for point estimation.

Uploaded by

Poonam Naidu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
71 views

General Concepts of Point Estimation

1. A point estimator is a single value that represents an estimate of an unknown parameter based on sample data. An unbiased estimator has an expected value equal to the true value of the parameter being estimated. 2. The maximum likelihood estimator is the value of the parameter(s) that maximizes the likelihood function, which describes the probability of obtaining the sample data given different possible parameter values. 3. Desirably, point estimators should be unbiased and have minimum variance; the maximum likelihood estimator often has these desirable properties, making it a commonly used method for point estimation.

Uploaded by

Poonam Naidu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Point Estimator Unbiased Estimator

6.1
Chapter 6 A point estimator of a parameter θ is a A point estimator θˆ is said to be an
single number that can be regarded as a unbiased estimator of θ if E (θˆ) = θ
General Concepts sensible value for θ . A point estimator for every possible value of θ . If θˆ is
Point Estimation of can be obtained by selecting a suitable biased, the difference E (θˆ) − θ is called
statistic and computing its value from the the bias of θˆ .
Point Estimation given sample data.

The pdf’s of a biased estimatorθˆ1and The pdf’s of a biased estimator θˆ1


Unbiased Estimator Principle of Unbiased Estimation
an unbiased estimator θˆ2 for a and an unbiased estimator θˆ2 for a
parameterθ . parameterθ .
pdf of θˆ2
pdf of θˆ2 When X is a binomial rv with When choosing among several different
pdf of θˆ1 parameters n and p, the sample estimators of θ , select one that is
pdf of θˆ1 proportion pˆ = X / n is an unbiased unbiased.
estimator of p.
θ θ
Bias of θ1
Bias of θ1

Unbiased Estimator Unbiased Estimator Principle of Minimum Variance Graphs of the pdf’s of two
Unbiased Estimation different unbiased estimators
Let X1, X2,…,Xn be a random sample If X1, X2,…,Xn is a random sample from a
from a distribution with mean µ and distribution with mean µ , then X is an Among all estimators of θ that are pdf of θˆ1
varianceσ 2. Then the estimator unbiased estimator of µ . If in addition unbiased, choose the one that has the
(X − X) minimum variance. The resulting θˆ is
2
the distribution is continuous and
i
symmetric, then the sample median and called the minimum variance unbiased pdf of θˆ2
σˆ = S =
2 2

n −1 any trimmed mean are also unbiased estimator (MVUE) of θ.


is an unbiased estimator. estimators of µ . θ

MVUE for a Normal Distribution A biased estimator that is preferable Standard Error
to the MVUE 6.2
Let X1, X2,…,Xn be a random sample The standard error of an estimator θˆ is
pdf of θˆ1 (biased)
from a normal distribution with its standard deviation σ θˆ = V (θˆ) . If Methods of
parameters µ and σ . Then the pdf of θˆ2 the standard error itself involves
estimator µˆ = X is the MVUE for µ .
(the MVUE) unknown parameters whose values can Point Estimation
be estimated, substitution into σ θˆ yields
the estimated standard error of the
θ estimator, denoted σˆθˆ or sθˆ .
Moment Estimators Likelihood Function Maximum Likelihood Estimators
Moments Let X1, X2,…,Xn be a random sample
Let X1, X2,…,Xn have joint pmf or pdf The maximum likelihood estimates
from a distribution with pmf or pdf
Let X1, X2,…,Xn be a random sample f ( x1 ,..., xn ;θ1 ,...,θ m ) (mle’s) θˆ1 ,...,θˆm are those values of the
f ( x;θ1 ,...,θ m ), where θ1 ,...,θ m
from a pmf or pdf f (x). For k = 1, 2,… are parameters whose values are where parameters θ1 ,...,θ m θi'
s that maximize the likelihood
the kth population moment, or kth unknown. Then the moment estimators have unknown values. When x1,…,xn are function so that
k
moment of the distribution f (x) is E ( X ). f ( x1 ,..., xn ;θˆ1 ,...,θˆm ) ≥ f ( x1 ,..., xn ;θ1 ,...,θ m )
θ1 ,...,θ m are obtained by equating the the observed sample values and f is
The kth sample moment is first m sample moments to the regarded as a function of θ1 ,...,θ m ,it is for all θ1 ,...,θ m
1 n corresponding first m population called the likelihood function. When the Xi’s are substituted in the place of
X k.
n i =1 i moments and solving for θ1 ,...,θ m . the xi’s, the maximum likelihood estimators
result.

The Invariance Principle Desirable Property of the Maximum


Likelihood Estimate
Let θˆ1 ,...,θˆm be the mle’s of the Under very general conditions on the joint
parameters θ1 ,...,θ m Then the mle of distribution of the sample, when the sample
any function h(θ1 ,...,θ m ) of these size n is large, the maximum likelihood
parameters is the function h(θˆ1 ,...,θˆm ) estimator of any parameter θ is approx.
of the mle’s.
unbiased [ E (θˆ) ≈ θ ] and has variance
that is nearly as small as can be achieved
by any estimator.
mleθˆ ≈ MVUE of θ

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy