masking data plots for graph

I want my plot to include my mask:

mask=true(size(x));

mask(0>x)=false;

mask=true(size(y));

mask(0>y)=false;

plot(x(mask),y(mask),'.')

It is not a problem, when I only have to mask the x values or y values seperately, only when I have to plot both masks.

Could the problem be that my mask for x and y respectively aren't having a particular definition? I tried with 'xmask' and 'ymask' and add this to the plotting sequence, but it does not work either.

Answer 1

[Walter Roberson]

xt = x;
xt(xt<0) = nan;
yt = y;
yt(yt<0) = nan;
plot(xt, yt);

[cvbnm]

I want to fit these data but my fit keeps just following the x-axis at x=0.

Is this the correct code?

hold on
[b,stats]=robustfit(xt,yt);

x1=[min(xt),max(xt)];

y1= b(1)+b(2)*x1;

a=plot(x1,y1,'-r');

[Walter Roberson]

mask = (x > 0) & (y > 0);
[b,stats]=robustfit(x(mask),y(mask));

However, this is only going to work if x and y are vectors. If they are 2D then the result of x(mask) and y(mask) is going to be a column vector, because 2D arrays cannot have "holes" in them.

[cvbnm]

They are not vectors, so that is maybe why it doesn't work, unfortunately. I added my original code below, just in case my reply to you would be confusing. Does this mean that I can not fit data, with I mask values for both the x- and y-axis?

Answer 2

[cvbnm]

My original code for fitting was:

[b,stats]=robustfit(x(mask),y(mask));

x1=[min(x(mask)),max(x(mask))];

y1= b(1)+b(2)*x1;

hold on

plot(x(mask),y(mask),'.b'); 

plot(x1,y1,'-r');

This is used for plotting my data and the fitting. When I use both of the two masks, it can not fit my data. There is no problem when I only masks one of the variables (either x or y).