Scribd
Upload
8 views19 pages

Understanding Point Estimation Methods

This document discusses point estimation and methods for estimating parameters from sample data. It introduces key concepts like population versus sample, parameters, and estimators. It describes two general methods for finding point estimators: the method of moments estimator (MME) and the maximum likelihood estimator (MLE). The MME matches sample moments to population moments, while the MLE chooses parameters that make the observed sample most likely. The document provides examples of applying these methods to estimate parameters of distributions like the gamma distribution from sample data. It also discusses evaluating estimators based on properties like consistency, efficiency, and invariance.

Uploaded by

Eva Chung
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
8 views19 pages

Understanding Point Estimation Methods

This document discusses point estimation and methods for estimating parameters from sample data. It introduces key concepts like population versus sample, parameters, and estimators. It describes two general methods for finding point estimators: the method of moments estimator (MME) and the maximum likelihood estimator (MLE). The MME matches sample moments to population moments, while the MLE chooses parameters that make the observed sample most likely. The document provides examples of applying these methods to estimate parameters of distributions like the gamma distribution from sample data. It also discusses evaluating estimators based on properties like consistency, efficiency, and invariance.

Uploaded by

Eva Chung
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

Chapter

7. Point Es1ma1on

1
Popula1on versus sample

!  Population: The entire Sample: The part of the
group of individuals in popula1on that we actually
which we are interested, examine, and for which we
but can’t usually assess do have data.
directly.
How well the sample
!  Aparameter is a number represents the popula1on
describing a depends on the sample
characteristic of the design.
Population
population.
Sample

2
Random sample

3
Es1mators

Simple example of an es1mator


The sample mean:






The popula1on mean: E[X]


4
Es1mators

If we assume the components of the data vector to be
i.i.d., generated from some underlying distribu1on,
then the popula1on mean is just E[X].

We can say that the sample mean is an es1mator of the
popula1on mean.

Roughly speaking, an es1mator is a sta1s1c which
“approximates” the quan1ty of interest.

5
Es1mators

1. Two general methods to find a point es1mator

Method of Moments Es1mator (MME)

Maximum Likelihood Es1mator (MLE)




6
Method of moments

7
Method of moments


8
Method of moments

Reminder: Gamma distribu1on




9
Method of moments


10
Method of moments

Example: Given a random sample of size n from a
Gamma distribu1on, es1mate the two parameters
by MME.


11
Maximum likelihood es1mator
Maximum Likelihood Es1mator (MLE)



12
Maximum likelihood es1mator

13
Maximum likelihood es1mator

Invariance Property of MLE:

Normal distribution

14
Maximum likelihood es1mator

15
Evalua1on of es1mators

2. Evalua1on of Es1mators

16
Evalua1on of es1mators

17
Evalua1on of es1mators

18
Evalua1on of es1mators

19

You might also like