Clear Filters
Clear Filters

how do i deduce the function using linear regression for a set of x and y values?

2 views (last 30 days)
clc
clear all
load x2.txt
load y2.txt
x=[x2]
y=log([y2])
format long
b2=x\y
yCalc1 = b2*x;
scatter(x,y)
hold on
plot(x,yCalc1)
xlabel('X_2')
ylabel('Y_2')
title('Linear Regression Relation Between X2 & Y2')
x = 25×1
70 170 10 20 100 180 200 110 130 31
y = 25×1
6.7153 8.8063 2.6804 3.7078 7.4083 8.8500 9.1882 7.6424 8.1274 4.9574
b2 =
0.050110078623913
This is what I am getting when i tried to use linear regression. Is there any way i can find the function this plot is tracing?

Accepted Answer

Turlough Hughes
Turlough Hughes on 2 Jan 2022
Consider using a power fit.
Your data:
x = readmatrix('https://uk.mathworks.com/matlabcentral/answers/uploaded_files/850570/x2.txt');
y = readmatrix('https://uk.mathworks.com/matlabcentral/answers/uploaded_files/850575/y2.txt');
Power law fit:
powerFit = fit(x, y, fittype('power1'))
powerFit =
General model Power1: powerFit(x) = a*x^b Coefficients (with 95% confidence bounds): a = 0.01469 (0.001567, 0.02782) b = 2.525 (2.359, 2.691)
plot(powerFit, x, y);
set(gca,'YScale','log')
  1 Comment
Turlough Hughes
Turlough Hughes on 2 Jan 2022
You actually have the parameters in your question but the way you fitted the data fixes the intercept to 0 - so the slope is equal to b2 and the intercept is 0. As a point of information, you can fit the slope and intercept using matrix left division with the following modification:
x = readmatrix('https://uk.mathworks.com/matlabcentral/answers/uploaded_files/850570/x2.txt');
y = readmatrix('https://uk.mathworks.com/matlabcentral/answers/uploaded_files/850575/y2.txt');
log_y = log(y);
b = [ones(size(x)) x]\log_y; % pad the left side with ones
yCalc1 = b(2)*x + b(1);
scatter(x,log_y)
hold on
plot(x,yCalc1)
sprintf('Slope: %1.2f\nIntercept: %1.2f',b(2),b(1))
ans =
'Slope: 0.02 Intercept: 4.45'

Sign in to comment.

More Answers (1)

KSSV
KSSV on 2 Jan 2022
clc
clear all
load x2.txt
load y2.txt
x=x2 ;
y = log(y2) ;
% Use polyfit
p = polyfit(x,y,1) ;
yCalc1 = polyval(p,x) ;
scatter(x,y)
hold on
plot(x,yCalc1)
title(sprintf('y = %f*x+%f',p))
xlabel('X_2')
ylabel('Y_2')

Categories

Find more on Interpolation in Help Center and File Exchange

Products


Release

R2021b

Community Treasure Hunt

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

Start Hunting!