w2preciprate::PairedMLE Class Reference

List of all members.

---

Detailed Description

This class can be used to provide maximum likelihood estimates if you have two pairs of estimates of a single variable.

We assume that both the estimates have non-zero variance, but that only the first estimate has a bias.

Author:

Lakshman

Version:

Id

w2preciprate_PairedMLE.h,v 1.3 2009/06/02 20:18:01 lakshman Exp

Public Member Functions
	PairedMLE ()
float	update (float x1, float var1, float x2, float var2)
	update with the pairs of observations x1 and x2, along with the variance of the two estimates.
float	getMLE (float x1) const
	get the maximum likelihood estimate of x if you only have x1
float	getBias () const
	get the bias of the first estimate.
bool	isValid () const
	Has this been updated at all?

---

Constructor & Destructor Documentation

w2preciprate::PairedMLE::PairedMLE

(

)

---

Member Function Documentation

float w2preciprate::PairedMLE::getBias

(

)

const [inline]

get the bias of the first estimate.

The second estimate is assumed to be unbiased.

float w2preciprate::PairedMLE::getMLE

(

float

x1

)

const [inline]

get the maximum likelihood estimate of x if you only have x1

bool w2preciprate::PairedMLE::isValid

(

)

const [inline]

Has this been updated at all?

float w2preciprate::PairedMLE::update	(	float	x1,
		float	var1,
		float	x2,
		float	var2
	)

update with the pairs of observations x1 and x2, along with the variance of the two estimates.

Remember that the second estimate should be unbiased.

Returns:

best estimate of x.

---