QUESTION 1 In nonlinear regression we use the builtin functi

QUESTION 1

In nonlinear regression we use the built-in function, fminsearch to minimize:

The distance between the data and the assumed curve.

The square of distance between the data and the assumed curve (S_r).

The square of distance between the data and the data mean (St).

The unknown coefficients

UESTION 3

Consider the following curve with unknown coefficients, a:

f(x)= a₀+a₁log(x)+a₂x²

The basis function matrix may be represented in MATLAB as:

Z=[a(1),a(2),a(3)]

Z=[1,x,x.^2]

Z=[ones(size(x)), log(x), x.^2]

Z=[ones(size(x)), x, x.^2]

QUESTION 5

The following command is used to determine the unknown coefficients, a:

a = fminsearch(@fSSR, [1,2,2], [], x, y)

From this command it is apparent that:

There are three values of both x and y.

There is one unkown coefficient.

There are two unkown coefficients.

There are three unkown coefficients.

Solution

Question 1 : In nonlinear regression we use the built-in function, fminsearch to minimize:

Answer is (a) : The distance between the data and the assumed curve.

Explaination : fminsearch is used to minimise norm of residual. To minimise the residual, one need to find the error in regression. which is the distance between the data and the assumed curve.