Help-Plot of two vectors - scatter plot with two subsamples??

Hi all,

I have the following question:

I have two column vectors X (min value= -0.5; max value= 1.1; 1000 observations) Y (min value =-0.7; max value= 3; 1000 observations)

I would like to perform the following plot. Plot a subset of X (for -0.1 to 0.5) on the x-axis and the corresponding observations of Y on the y-axis.

I would like to split the Y observations in two groups around the point X=0, such that I have observations to the left of X=0 and to the right of X=0.I would also like to fit regression lines (lines of best fit) to the left and right of X=0 and compare their slopes. Can anybody help me with this difficult challenge?

Thanks for your time!

Answer 1

[Walter Roberson]

availcolors = [1 0 0; 0 1 0];  %arbitrary, red and green
colidx = 1 + (X >= 0);    %1 on the left, 2 on the right
colorarray = availcolors(colidx,:);  %array of rgb
pointsize = 10;  %arbitrary
scatter(X, Y, pointsize, colorarray)

Answer 2

[Jarek]

Thank you Walter Roberson for taking the time to look through my problem.

That is great!

Still, two issues remaining:

1. How to plot a subset of X (for -0.1 to 0.5) on the x-axis and the corresponding observations of Y on the y-axis.

2. I would also like to fit regression lines (lines of best fit) to the left and right of X=0 and compare their slopes.

It would be great if somebody could help me on this?

Thanks for your time and effort!

[Walter Roberson]

in_subset = (X >= -0.1 & X <= 0.5);

X = X(in_subset);

Y = Y(in_subset);

Regression lines you should now be able to figure out for yourself using logical indexing.

[Jarek]

Thank you Walter Roberson for the swift reply, this is really cool!!

As you can see, I am completely new to matlab.

I have found the Basic Fitting sub-menu in Tools-menu, however it fits only one line for the whole sample, and I need two separate lines for X=[-0.1 to 0] and [0 to 0.5].

Unfortunately, I couldn’t figure out how this logical indexing works, still trying though.

It could be great if you could help on this!

Thanks again for your time and effort!