Statistical Estimation Website Links For
Theory 		 

Information About

Statistical Estimation
  CATEGORIES ABOUT ESTIMATION THEORY
   
estimation theory
   
signal processingestimation theory
   
signal processing
   
information theory
   
signal processing
   
statistical theory
   
APPAREL
BABY
BEAUTY
BOOKS
CAR TOYS
CELL PHONES
DVD'S
ELECTRONICS
GOURMET FOOD
GROCERIES
HEALTH & PERSONAL
HOME & GARDEN
JEWELRY
MUSIC
MUSIC INSTRUMENTS
OFFICE PRODUCTS
SOFTWARE
SPORTING GOODS
TOOLS & HARDWARE
TOYS
VIDEO GAMES
SHOPPING HOME
MORE SHOPPING...



For example, it is desired to estimate the proportion of a population of voters who will vote for a particular candidate. That proportion is the unobservable parameter; the estimate is based on a small random sample of voters.

Or, for example, in Radar the goal is to estimate the location of objects (airplanes, boats, etc.) by analyzing the received echo and a possible question to be posed is "where are the airplanes?"
To answer where the airplanes are, it is necessary to estimate the distance the airplanes are at from the radar station, which can provide an absolute location if the absolute location of the radar station is known.

In estimation theory, it is assumed that the desired information is embedded into a Noisy Signal .
Noise adds uncertainty and if there was no uncertainty then there would be no need for estimation.


FIELDS THAT USE ESTIMATION THEORY

There are numerous fields that require the use of estimation theory.
Some of these fields include (but by no means limited to):



The measured data is likely to be subject to Noise or uncertainty and it is through statistical Probability that Optimal solutions are sought to extract as much Information from the data.


ESTIMATION PROCESS

The entire purpose of estimation theory is to arrive at an estimator, and preferably an implementable one that could actually be used.
The estimator takes the measured data as input and produces an estimate of the parameters.

It is also preferable to derive an estimator that exhibits Optimality .
An optimal estimator would indicate that all available information in the measured data has been extracted, for if there was unused information in the data then the estimator would not be optimal.

These are the general steps to arrive at an estimator:
  • In order to arrive at a desired estimator for estimating a single or multiple parameters, it is first necessary to determine a model for the system. This model should incorporate the process being modeled as well as points of uncertainty and noise. The model describes the physical scenario in which the parameters apply.

  • After deciding upon a model, it is helpful to find the limitations placed upon an estimator. This limitation, for example, can be found through the Cramér-Rao Inequality .

  • Next, an estimator needs to be developed or applied if an already known estimator is valid for the model. The estimator needs to be tested against the limitations to determine if it is an optimal estimator (if so, then no other estimator will perform better).

  • Finally, experiments or simulations can be ran with the estimator to test the performance.


After arriving at an estimator, real data might show that the model used to derive the estimator is incorrect, which may require repeating these steps to find a new estimator.
A non-implementable or infeasible estimator may need to be scrapped and the process start anew.

In summary, the estimator estimates the parameters of a physical model based on measured data.


BASICS

To build a model, several statistical "ingredients" need to be known.
These are needed to ensure the estimator has some mathematical tractability instead of being based on "gut feel."

The first is a set of Statistical Sample s taken from a Random Vector (RV) of size N. Put into a Vector ,

: \mathbf{x} = \begin{bmatrix} x \ x[1 \ dots \ x[N-1] \end{bmatrix}.

Secondly, we have the corresponding M parameters

: \mathbf{ heta} = \begin{bmatrix} heta_1 \ heta_2 \ dots \ heta_M \end{bmatrix},

which need to be established with their Probability Density Function (pdf) or Probability Mass Function (pmf)



One common estimator is the Minimum Mean Squared Error (MMSE) estimator, which utilizes the error between the estimated parameters and the actual value of the parameters

: \mathbf{e} = \hat{\mathbf{ heta}} - \mathbf{ heta}

as the basis for optimality.
This error term is then squared and minimized for the MMSE estimator.


ESTIMATORS

This list is some of the more common Estimator s used, and some topics related to them:


EXAMPLE: DC GAIN IN WHITE GAUSSIAN NOISE

Consider a received Discrete Signal , x of N Independent Samples that consists of a DC gain A with Additive White Gaussian Noise w[n with ''known'' Variance \sigma^2 (i.e., \mathcal{N}(0, \sigma^2)).
Since the variance is known then the only unknown parameter is A.

The model for the signal is then
: x = A + w[n \quad n=0, 1, \dots, N-1

Two possible (of many) estimators are:

Both of these estimators have a Mean of A, which can be shown through taking the Expected Value of each estimator

:\mathrm{E}\left[\hat{A}_1 ight] = \mathrm{E}\left x[0 ight] = A
and
:
\mathrm{E}\left \hat{A}_2 ight
=
\mathrm{E}\left[ rac{1}{N} \sum_{n=0}^{N-1} x[n] ight]
=
rac{1}{N} \left \sum_{n=0}^{N-1} \mathrm{E}\left[ x[n ight] ight]
=
rac{1}{N} \left N A ight
=
A


At this point, these two estimators would appear to perform the same.
However, the difference between them becomes apparent when comparing the variances.

:\mathrm{var} \left( \hat{A}_1 ight) = \mathrm{var} \left( x {Link without Title} ight) = \sigma^2
and
:
\mathrm{var} \left( \hat{A}_2 ight)
=
\mathrm{var} \left( rac{1}{N} \sum_{n=0}^{N-1} x {Link without Title} ight)
=
rac{1}{N^2} \left \sum_{n=0}^{N-1} \mathrm{var} (x[n ) ight]
=
rac{1}{N^2} \left N \sigma^2 ight
=
rac{\sigma^2}{N}


It would seem that the sample mean is a better estimator since, as N o \infty, the variance goes to zero.


Maximum likelihood

Continuing the example using the Maximum Likelihood estimator, the Probability Density Function (pdf) of the noise for one sample w {Link without Title} is

:p(w = rac{1}{\sigma \sqrt{2 \pi}} \exp\left(- rac{1}{2 \sigma^2} w[n ^2 ight)

and the probability of x becomes (x[n can be thought of a \mathcal{N}(A, \sigma^2))

:p(x A) = rac{1}{\sigma \sqrt{2 \pi}} \exp\left(- rac{1}{2 \sigma^2} (x[n - A)^2 ight)

By independence, the probability of \mathbf{x} becomes

:
p(\mathbf{x}; A)
=
\prod_{n=0}^{N-1} p(x {Link without Title} ; A)
=
rac{1}{\left(\sigma \sqrt{2\pi} ight)^N}
\exp\left(- rac{1}{2 \sigma^2} \sum_{n=0}^{N-1}(x {Link without Title} - A)^2 ight)


Taking the Natural Logarithm of the pdf

:
\ln p(\mathbf{x}; A)
=
-N \ln \left(\sigma \sqrt{2\pi} ight)
- rac{1}{2 \sigma^2} \sum_{n=0}^{N-1}(x {Link without Title} - A)^2


and the maximum likelihood estimator is

:\hat{A} = \arg \max \ln p(\mathbf{x}; A)

Taking the first Derivative of the log-likelihood function

:
rac{\partial}{\partial A} \ln p(\mathbf{x}; A)
=
rac{1}{\sigma^2} \left \sum_{n=0}^{N-1}(x[n - A) ight]
=
rac{1}{\sigma^2} \left \sum_{n=0}^{N-1}x[n - N A ight]


and setting it to zero

:
0
=
rac{1}{\sigma^2} \left \sum_{n=0}^{N-1}x[n - N A ight]
=
\sum_{n=0}^{N-1}x {Link without Title} - N A


This results in the maximum likelihood estimator

:
\hat{A} = rac{1}{N} \sum_{n=0}^{N-1}x {Link without Title}


which is simply the sample mean.
From this example, it was found that the sample mean is the maximum likelihood estimator for N samples of AWGN with a fixed, unknown DC gain.


Cramér-Rao lower bounds

To find the Cramér-Rao Lower Bounds (CRLB) of the sample mean estimator, it is first necessary to find the Fisher Information number

:
\mathcal{I}(A)
=
\mathrm{E}
\left(
\left[
rac{\partial}{\partial heta} \ln p(\mathbf{x}; A)
ight]^2
ight)
=
-\mathrm{E}
\left[
rac{\partial^2}{\partial heta^2} \ln p(\mathbf{x}; A)
ight]


and copying from above

:
rac{\partial}{\partial A} \ln p(\mathbf{x}; A)
=
rac{1}{\sigma^2} \left \sum_{n=0}^{N-1}x[n - N A ight]


Taking the second derivative
:
rac{\partial^2}{\partial A^2} \ln p(\mathbf{x}; A)
=
rac{1}{\sigma^2} (- N)
=
rac{-N}{\sigma^2}


and finding the negative expected value is trivial since it is now a deterministic constant

-\mathrm{E}
\left[
rac{\partial^2}{\partial A^2} \ln p(\mathbf{x}; A)
ight]
=
rac{N}{\sigma^2}


Finally, putting the Fisher information into

:
\mathrm{var}\left( \hat{A} ight)
\geq
rac{1}{\mathcal{I}}


results in

:
\mathrm{var}\left( \hat{A} ight)
\geq
rac{\sigma^2}{N}


Comparing this to the variance of the sample mean (determined previously) shows that the sample mean is ''equal to'' the Cramér-Rao lower bounds for all values of N and A.
The sample mean is the Minimum Variance Unbiased Estimator (MVUE) in addition to being the Maximum Likelihood estimator.

This example of DC gain + WGN is a recurring example in Kay's ''Fundamentals of Statistical Signal Processing''.


BOOKS



SEE ALSO