Histogram problem

I plotted a histogram. I now trying to chang the way the x- axis is read.

The Data is huge so it currently reads 0, 200, 400, 800, 1000 Meanwhile i would like it to show the other bin sizes. my bin width is 5, therefore

I would like it to show 0, 5, 10, 15, 20..... 1000

I tried this hist(data(1:29196,1),[0:5:1000]);figure(gcf);

Thanks for any help received

Answer 1

[Walter Roberson]

set(gca, 'XTick', 0:5:1000)

However, this seems likely to want to place the labels so closely together that they will become unreadable.

If your data is sufficiently large, there is a possibility that representing it all would require more space than you have available in your figure. There really isn't any effective way of dealing with that except for allowing more space for the axes, or providing a partially-visible uipanel with scroll bars (FEX contributions available) that shows the image at a higher resolution.

Answer 2

[the cyclist]

It would be helpful if you posted the code that generates the histogram, but I am guessing what you are asking for is this:

set(gca,'XTick',0:5:1000)

However, I don't think so many ticks are going to be legible. You might need to compromise.

[Anthony]

hist(data(1:29196,1),[0:5:1000]);figure(gcf);

Would like to know where exactly to place set(gca,'XTick',0:5:1000)

[Walter Roberson]

Right after the hist() call.

You should not need to figure() call unless have have done something unusual such as set(0,'CurrentFigure')

[Walter Roberson]

It looks to me as if probably you should instead be using

[b,bin] = histc(data(1:29196,1),0:5:1000);

bar(bin, b);

after that you might find you do not need to set the xtick.

[Anthony]

I tried this and this is the error i get

[b,bin] = histc(data(1:29196,1),0:5:1000);

bar(bin, b);

??? Error using ==> bar at 57

X must be same length as Y.

[the cyclist]

Try it with the hist() function rather than the histc() function (which creates an extra bin because of how it handles edges). [You might want to define your bin centers as 2.5:5:997.5.]

[Kelly Kearney]

Actually, I think Walter is misusing the outputs of histc:

edge = 0:5:1000;

[b,bin] = histc(data(1:29196,1), edge);

bar(edge,b,'histc');

is what I think he intended.

[Walter Roberson]

Yes, that would make sense, Kelly. This could be further reduced to

edge = 0:5:1000;

bar(edge, histc(data(1:29196,1),edge), 'histc')