You get a bonus - 1 coin for daily activity. Now you have 1 coin

Likelihood Maximum Method

Lecture



The likelihood function introduced by Fisher looks like this:

  Likelihood Maximum Method

Where   Likelihood Maximum Method - unknown parameter.

As a parameter estimate   Likelihood Maximum Method need to choose a value at which   Likelihood Maximum Method reaches a maximum. Insofar as   Likelihood Maximum Method reaches a maximum with the same   Likelihood Maximum Method what and   Likelihood Maximum Method , then search for the required value of the estimate   Likelihood Maximum Method consists in solving the likelihood equation   Likelihood Maximum Method . At the same time all the roots   Likelihood Maximum Method should be discarded, and leave only those that depend on   Likelihood Maximum Method .

The estimate of the distribution parameter is a random variable that has a mathematical expectation and “scattering” around it. An estimate is said to be effective if its “scattering” around its expectation is minimal.

The following theorem is valid (given without proof). If exists for   Likelihood Maximum Method effective evaluation   Likelihood Maximum Method , then the likelihood equation has a unique solution. This decision   Likelihood Maximum Method converges to the true value   Likelihood Maximum Method .

All this is true for several unknown parameters. For example, for a one-dimensional normal law

  Likelihood Maximum Method ,

  Likelihood Maximum Method ,

from here   Likelihood Maximum Method at   Likelihood Maximum Method .

  Likelihood Maximum Method ,

from here   Likelihood Maximum Method .

The estimate is called unbiased if the expectation   Likelihood Maximum Method ratings   Likelihood Maximum Method equally   Likelihood Maximum Method . Evaluation   Likelihood Maximum Method is unbiased. Indeed, since   Likelihood Maximum Method - simple random selection from the general population, then   Likelihood Maximum Method and   Likelihood Maximum Method .

Find out if   Likelihood Maximum Method obtained by the maximum likelihood method (or the method of moments), unbiased. Easy to make sure that

  Likelihood Maximum Method .

Consequently,

  Likelihood Maximum Method

  Likelihood Maximum Method .

Find the expected value of this value:

  Likelihood Maximum Method .

Since the variance   Likelihood Maximum Method independent of value   Likelihood Maximum Method then choose   Likelihood Maximum Method . Then

  Likelihood Maximum Method ,   Likelihood Maximum Method ,   Likelihood Maximum Method ,

Where   Likelihood Maximum Method ,   Likelihood Maximum Method - correlation coefficient between   Likelihood Maximum Method and   Likelihood Maximum Method (in this case it is equal to zero, since   Likelihood Maximum Method and   Likelihood Maximum Method do not depend on each other).

So,   Likelihood Maximum Method . This shows that the estimate   Likelihood Maximum Method is not unbiased, its expectation is somewhat less than   Likelihood Maximum Method . To eliminate this bias you need to multiply   Likelihood Maximum Method on   Likelihood Maximum Method . As a result, we get an unbiased estimate.

  Likelihood Maximum Method .


Comments


To leave a comment
If you have any suggestion, idea, thanks or comment, feel free to write. We really value feedback and are glad to hear your opinion.
To reply

Pattern recognition

Terms: Pattern recognition