How can I rectify this error? 'Invalid use of Operator.' 'Too many Output arguements'

affineTransformGauss calculates the mean and covariance of y, the transformed variable, exactly when the function, f, is defined as y = f(x) = Ax + b, where A is a matrix, b is a vector of the same dimensions as y, and x is a Gaussian random variable.

%Input
%   MU_X        [n x 1] Expected value of x.
%   SIGMA_X     [n x n] Covariance of x.
%   A           [m x n] Linear transform matrix.
%   B           [m x 1] Constant part of the affine transformation.
%   x           
%Output
%   MU_Y        [m x 1] Expected value of y.
%   SIGMA_Y     [m x m] Covariance of y.
function [mu_y, Sigma_y] = affineGaussianTransform(mu_x, Sigma_x, A, b)
%Code here
x=horzcat(mu_x,Sigma_x);
C=A*x;
[mu_y, Sigma_y]= (C+b) 
end
%Code to call your function
mu_x= [7;3];
Sigma_x= [0.2 0; 0 8];
A=[1 1; 1 -1];
b=0;
%Calculate y
[mu_y, Sigma_y]=affineGaussianTransform(mu_x, Sigma_x, A, b)

[madhan ravi]

[mu_y, Sigma_y]= (C+b) 

Returns single output whereas you have declared two

[Romy Jemals]

I need [mu_y, Sigma_y] such that mu_y & Sigma_y are matrixes of order m*1 and m*m through 'Ax+b' where A is a matrix of m*n and B is a vector of m*1. How shall I correct the mistake?

[Stephen23]

This line will throw an error :

[mu_y, Sigma_y]= (C+b)

The addition operator takes two arrays and returns one array, exactly as its documentation shows. The addition operator does NOT return two arrays, as you are using it.

What do you expect its two outputs to be?

[Romy Jemals]

I need [mu_y, Sigma_y] such that

mu_y & Sigma_y are matrixes of order m*1 and m*m ;through 'xy=Ax+b' where

A is a matrix of m*n,

x is [mu_x, Sigma_y] mu_x is n*1 and Sigma_x is n*n.

and B is a vector of m*1.

How shall I correct the mistake?

[madhan ravi]

Perhaps all you need is

doc deal % ?

Answer 1

[VBBV]

%Input
%   MU_X        [n x 1] Expected value of x.
%   SIGMA_X     [n x n] Covariance of x.
%   A           [m x n] Linear transform matrix.
%   B           [m x 1] Constant part of the affine transformation.
%   x           
%Output
%   MU_Y        [m x 1] Expected value of y.
%   SIGMA_Y     [m x m] Covariance of y.

%Code to call your function
mu_x= [7;3];
Sigma_x= [0.2 0; 0 8];
A=[1 1; 1 -1];
b=zeros(length(A),1);  % Make B as vector 
%Calculate y
[mu_y, Sigma_y]=affineGaussianTransform(mu_x, Sigma_x, A, b)

mu_y = 3×1

3.3333 0.0667 2.6667

Sigma_y = 2×3

10.0000 0.2000 8.0000 4.0000 0.2000 -8.0000

function [mu_y, Sigma_y] = affineGaussianTransform(mu_x, Sigma_x, A, b)
%Code here
x=horzcat(mu_x,Sigma_x);
C=A.'*x+b;  % calculate mean 
Sigma_y = C;
mu_y = (sum(x)./length(Sigma_y)).'; 
end

Define the B vector as

b=zeros(length(A),1); 

and compute the mean as

mu_y = (sum(x)./length(Sigma_y)).';