##### Page tree
Go to start of banner

# Linear regression

You are viewing an old version of this page. View the current version.

Version 9

## Problem definition

In speech processing and elsewhere, a frequently appearing task is to make a prediction of an unknown vector y from available observation vectors x. Specifically, we want to have an estimate $$\hat y = f(x)$$ such that $$\hat y \approx y.$$ In particular, we will focus on linear estimates where $$\hat y=f(x):=A^T x,$$ and where A is a matrix of parameters.

## The minimum mean square estimate (MMSE)

Suppose we want to minimise the squared error of our estimate on average. The estimation error is $$e=y-\hat y$$ and the squared error is the L2-norm of the error, that is, $$\left\|e\right\|^2 = e^T e$$ and its mean can be written as the expectation $$E\left[\left\|e\right\|^2\right] = E\left[\left\|y-\hat y\right\|^2\right] = E\left[\left\|y-A^T x\right\|^2\right].$$ Formally, the minimum mean square problem can then be written as

$\min_A\, E\left[\left\|y-A^T x\right\|^2\right].$

This can in generally not be directly implemented because we have the abstract expectation-operation in the middle.

(Advanced derivation) To get a computational model, first note that the error expectation can be written in terms of the mean of a sample of vector ek as

$E\left[\left\|e\right\|^2\right] \approx \frac1N \sum_{k=1}^N \left\|e_k\right\|^2 = \frac1N {\mathrm{tr}}(E^T E),$

where $$E=[e_1,\,e_2,\dotsc,e_N]$$ and tr() is the matrix trace. To minimize the error energy expectation, we can then set its derivative to zero

$0 = \frac{\partial}{\partial A} \frac1N {\mathrm{tr}}(E^T E) = \frac1N\frac{\partial}{\partial A} {\mathrm{tr}}((Y-A^TX)^T (Y-A^TX)) = \frac1N(Y-A^T X)X^T$

where the observation matrix is $$X=[x_1,\,x_2,\dotsc,x_N]$$ and the desired output matrix is $$Y=[y_1,\,y_2,\dotsc,y_N]$$ . (End of advanced derivatione)

It follows that the optimal weight matrix A can be solved as

$\boxed{A = (XX^T)^{-1}XY^T = X^\dagger Y^T},$

where the superscript $$\dagger$$ denotes the Moore-Penrose pseudo-inverse.

• No labels