In statistics, a location parameter of a probability distribution is a scalar- or vector-valued parameter , which determines the "location" or shift of the distribution. In the literature of location parameter estimation, the probability distributions with such parameter are found to be formally defined in one of the following equivalent ways:

A direct example of a location parameter is the parameter of the normal distribution. To see this, note that the probability density function of a normal distribution can have the parameter factored out and be written as:

thus fulfilling the first of the definitions given above.

The above definition indicates, in the one-dimensional case, that if is increased, the probability density or mass function shifts rigidly to the right, maintaining its exact shape.

A location parameter can also be found in families having more than one parameter, such as location–scale families. In this case, the probability density function or probability mass function will be a special case of the more general form

where is the location parameter, θ represents additional parameters, and is a function parametrized on the additional parameters.

Definition[4]

edit

Let   be any probability density function and let   and   be any given constants. Then the function

 

is a probability density function.


The location family is then defined as follows:

Let   be any probability density function. Then the family of probability density functions   is called the location family with standard probability density function  , where   is called the location parameter for the family.

Additive noise

edit

An alternative way of thinking of location families is through the concept of additive noise. If   is a constant and W is random noise with probability density   then   has probability density   and its distribution is therefore part of a location family.

Proofs

edit

For the continuous univariate case, consider a probability density function  , where   is a vector of parameters. A location parameter   can be added by defining:

 

it can be proved that   is a p.d.f. by verifying if it respects the two conditions[5]   and  .   integrates to 1 because:

 

now making the variable change   and updating the integration interval accordingly yields:

 

because   is a p.d.f. by hypothesis.   follows from   sharing the same image of  , which is a p.d.f. so its image is contained in  .

See also

edit

References

edit
  1. ^ Takeuchi, Kei (1971). "A Uniformly Asymptotically Efficient Estimator of a Location Parameter". Journal of the American Statistical Association. 66 (334): 292–301. doi:10.1080/01621459.1971.10482258. S2CID 120949417.
  2. ^ Huber, Peter J. (1992). "Robust Estimation of a Location Parameter". Breakthroughs in Statistics. Springer Series in Statistics. Springer. pp. 492–518. doi:10.1007/978-1-4612-4380-9_35. ISBN 978-0-387-94039-7.
  3. ^ Stone, Charles J. (1975). "Adaptive Maximum Likelihood Estimators of a Location Parameter". The Annals of Statistics. 3 (2): 267–284. doi:10.1214/aos/1176343056.
  4. ^ Casella, George; Berger, Roger (2001). Statistical Inference (2nd ed.). Thomson Learning. p. 116. ISBN 978-0534243128.
  5. ^ Ross, Sheldon (2010). Introduction to probability models. Amsterdam Boston: Academic Press. ISBN 978-0-12-375686-2. OCLC 444116127.