MATLAB Answers

Undefined function 'mtimes' for input arguments of type 'function_handle'.

9 views (last 30 days)
Liam Crocker
Liam Crocker on 26 Sep 2020
Commented: Liam Crocker on 26 Sep 2020
So I'm very puzzled as to how I find out my C values from the C lin fit. It tells me define my formula's as function handles, no worries done that easy. Then it tells me perform least squares fit or nlinfit of my X function (the Tib model for distance). So I've laid it out in my code the same way I've done it before and I get the error you see in the title. Now I'm confused because X relies on A, and A relies on C. I tried doing an nlinfit on A to find C but no luck there. So I'm stuck on part a where things are highlighted.
I read others with similar errors and the soulition seemed to be pass something through your function, which I tried by
c = nlinfit(x,t,X(c,t),c0);
but that didn't work either, just told me c was an unknown variable.
clc
clear
%V(t) = @(t) A*(1-exp(c(1)*(t-t0); %speed Keller Model
%X(t) = @(t) A*(t-t0)-(A/c(1))*(1-exp(-c(1)*(t-t0)); %distance Keller Model
V = @(c, t) A*(1-exp(-c(1)*K))-c(2).*t; %speed
X = @(c, t) A*K-(c(1)/2)*K^2-(A/c(2))*(1-exp(-c(2)*K)); %Tibshirani model for distance
%time => 0.1; %minimum reaction time
A = @(c) (xe+c(1)/2)*(K^2)/(K-((1/c(2))*1-exp(-c(2)*K))); % A parameter, kept here untill script runs well
c0 = [.01;1;];
%--------------------------------------------------------------------------------------------------------%
c0 = [.01; 1]; %initial c parameters
load sprint.dat
x = sprint(:,1);
tbolt1 = sprint(:,2); %Time_Bolt_Beijing
tbolt2= sprint(:,3); %Time_Bolt_Berlin
tlewis1 = sprint(:,4); %Time_C.Lewis_Rome
tjohn1= sprint(:,4); %Time_B.Johnson_Rome
t0 = tbolt1(1);
te = tbolt1(11); %end time for bolt's first run in bejing
t = tbolt1;
K = te-t0; %shorens equation for bolt
xe = 100; %Ending distances for all runners
A = @(c) (xe+c(1)/2)*(K^2)/(K-((1/c(2))*1-exp(-c(2)*K))); % A parameter
V = @(c, t) A*(1-exp(-c(1)*K))-c(2).*t; %speed
X = @(c, t) A*K-(c(1)/2)*K^2-(A/c(2))*(1-exp(-c(2)*K)); %Tibshirani model for distance
%--------------------------------------------------------------------------------------------------------%
c = nlinfit(x,t,X,c0);

  0 Comments

Sign in to comment.

Accepted Answer

Steven Lord
Steven Lord on 26 Sep 2020
You can't multiply a function handle and a number or two function handles.
You can multiply the result you get from evaluating a function handle and a number.
x = 0:360;
f = @sind;
g = @(x) 2*f(x);
figure('Name', 'g = @(x) 2*f(x)')
plot(x, g(x)) % works
h = @(x) 2*f;
figure('Name', 'h = @(x) 2*f')
plot(x, h(x))% does not work
You can multiply the values reurned by two function handles.
q = @(x) f(x).*g(x);
figure('Name', 'q = @(x) f(x).*g(x)')
plot(x, q(x))

  1 Comment

Liam Crocker
Liam Crocker on 26 Sep 2020
I understand this example thank you, but am unsure how I would fix my coding here sorry!.
I can't multiply my X by the function handle A, I need to be multiplying the result I get from this. But the result of A relies on the values of C, which I dont know. So I tried to find C using the nlinfit for A, to help nail down a result of A to use with my X formula. It returns me this error:
Error using nlinfit (line 213)
Error evaluating model function '@(c)(xe+c(1)/2)*(K^2)/(K-((1/c(2))*1-exp(-c(2)*K)))'.
Error in Assignment2Q3 (line 29)
c = nlinfit(x,t,A,c0);
Caused by:
Error using Assignment2Q3>@(c)(xe+c(1)/2)*(K^2)/(K-((1/c(2))*1-exp(-c(2)*K)))
Too many input arguments.

Sign in to comment.

More Answers (1)

David Hill
David Hill on 26 Sep 2020
Edited: David Hill on 26 Sep 2020
c = [.01; 1];
x = 0:10:100;
tbolt1 = [0.165,1.85,2.87,3.78,4.65,5.5,6.32,7.14,7.96,8.79,9.69];
t0 = tbolt1(1);
te = tbolt1(11);
t = tbolt1;
K = te-t0;
xe = 100;
A = @(c) (xe+c(1)/2*K^2)/(K-1/c(2)*(1-exp(-c(2)*K)));
V = @(c,t) A(c)*(1-exp(-c(2)*(t-t0)))-c(1)*t;
X = @(c, t) A(c)*(t-t0)-c(1)/2*(t-t0).^2-A(c)/c(2)*(1-exp(-c(2)*(t-t0)));
C= nlinfit(t,x,X,c);

Products


Release

R2020a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!