Scribd
Upload
352 views

Maximum-Likelihood Estimation

Maximum-likelihood estimation is a general method for estimating parameters in statistical models. It involves finding the parameter values that maximize the likelihood function, or the probability of obtaining the sample results given the parameters. The method can be applied to models with any probability distribution, not just normal distributions like least squares. For example, to estimate the probability p of individuals being infected by a parasite based on sample data, the likelihood function is defined as the probability of the data given p. The value of p that maximizes this likelihood function is the maximum likelihood estimate.

Uploaded by

kevinalmer
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
352 views

Maximum-Likelihood Estimation

Maximum-likelihood estimation is a general method for estimating parameters in statistical models. It involves finding the parameter values that maximize the likelihood function, or the probability of obtaining the sample results given the parameters. The method can be applied to models with any probability distribution, not just normal distributions like least squares. For example, to estimate the probability p of individuals being infected by a parasite based on sample data, the likelihood function is defined as the probability of the data given p. The value of p that maximizes this likelihood function is the maximum likelihood estimate.

Uploaded by

kevinalmer
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 6

Maximum-likelihood

estimation
A general method for
estimating parameters in a
model

The method of leastModel for the


squares

expectation (fixed part


of the model):
E[Y ] x
i

1 i

Residuals:ri yi E[Yi ]
The method of leastsquares:
Find
the values for the parameters (0 and
1) that makes the sum of the squared
residuals (rj2) as small as possible.

Can only be used when the


error term is normal (residuals
are assumed to be drawn from
a Ynormal
distribution)
x , where ~ N (0, )
i

1 i

Model for the


expectation (fixed part
of the model):
E[Y ] x
i

1 i

Residuals:ri yi E[Yi ]

The maximum
likelihood method
-is
Can
be applied
to models
more
general!
with any probability
distribution

The maximum likelihood


Example:
We want to estimate the probability, p, that
individuals are infected with a certain kind of
Probability
parasite.
The maximum likelihood
Ind.:

1
2

Infected of
observatio
: 1
p
n:

1-p

1-p

1-p

1-p

10

method (discrete
distribution):
1.Write down the probability
of each observation by using
the model parameters
2.Write down the probability
of all the
data4
6

Pr(Data | p) p (1 p )

3.Find the value

The maximum likelihood


Example:
We want to estimate the probability, p, that
individuals are infected with a certain kind of
Probability
parasite.
Likelihood function:

1-p

1-p

1-p

10

- Find the value parameter(s)


that maximize this
probability
0.0012

1-p

0.0008

0.0004

L( p) Pr( Data | p ) p 6 (1 p ) 4

0.0000

Infected of
observatio
: 1
p
n:

L(p, K, N)

Ind.:

0.0

0.2

0.4

0.6
p

0.8

1.0

Exercises!

You might also like