linear fit

When plotting a scatter plot is it possible to add a linear fit to the the graph without having to go into tools-> basic fitting and clicking on linear and show equations?

Answer 1

[Wayne King]

lsline is in the Statistics Toolbox, if you do not have that product you can use polyfit() to fit a 1st order polynomial.

    x = 1:10; 
    y1 = x + randn(1,10); 
    scatter(x,y1,25,'b','*') 
    P = polyfit(x,y1,1);
    yfit = P(1)*x+P(2);
    hold on;
    plot(x,yfit,'r-.');

[Richard]

great that works. And what about obtaining the equation for the straight line: y=mx+c?

[Wayne King]

yfit = P(1)*x+P(2)

is the equation.... P(1) is the slope and P(2) is the intercept

[Nikolaus Koopmann]

yfit=@(x)P(1)*x+P(2)

is what he means, i think

[Galina Machavariani]

Thank you for the answer.

And how can I display linear fit equation on the graph?

I mean not "manual" way in the graph editor, , but the code ...

Thank you !

[Seth DeLand]

You would need to create the string of the equation and then place it on the graph with "text". Here is an expanded version of Wayne's example that does this:

x = 1:10;
y1 = x + randn(1,10);
scatter(x,y1,25,'b','*')
P = polyfit(x,y1,1);
yfit = polyval(P,x);
hold on;
plot(x,yfit,'r-.');
eqn = string(" Linear: y = " + P(1)) + "x + " + string(P(2));
text(min(x),max(y1),eqn,"HorizontalAlignment","left","VerticalAlignment","top")

Answer 2

[Thomas]

Also you can always do it once manually, generate data set, create the plot, make the linear fit with the equations, then in the Figure window

File>Generate code..

This will create a MATLAB function for everything that you did manually and can use it again and again if you have more data sets.

[Galina Machavariani]

After I did linear fit with equation, What should I write in the command window to generate the code?

Answer 3

[Wayne King]

Hi, yes, you can use lsline()

x = 1:10;
y1 = x + randn(1,10);
scatter(x,y1,25,'b','*')
hold on;
lsline;

[Richard]

matlab doesnt seem to recognise isline! Am I missing a toolbox, the version I'm working with is 2011b?

[Wayne King]

lsline(), not isline. Did you copy and paste my code? I didn't write isline.

[Richard]

>> clear all

>> x = 1:10;

y1 = x + randn(1,10);

scatter(x,y1,25,'b','*')

hold on;

lsline;

Undefined function or variable 'lsline'.

[vkehayas]

The statistics toolbox is required for

lsline

Answer 4

[Annu Panwar]

but anyone has observed that the results are different when you do polyfit by using codes and manually?

[Danhay]

that is true. I observed that too

[Namira]

I observed that too. Do you know the solution?

Answer 5

[sabreen haj]

Can you help me to write script for calibration curve And give me the equation so i can finde the x value then the result shown in a table with everage of 3 x value and std

Answer 6

[Marcello Wienhoven]

x = 1:10;

y1 = x + randn(1,10);

scatter(x,y1,25,'b','*')

P = polyfit(x,y1,1);

yfit = P(1)*x+P(2);

hold on;

plot(x,yfit,'r-.');

[Galina Machavariani]

Thank you for the answer.

And how can I display linear fit equation on the graph?

I mean not "manual" way in the graph editor, , but the code ...

Thank you !