X Y scatter data Color Contour of Frequency Distribution spatially

I have X Y scatter data ranging from -125km to +125km in both X and Y.

In 5 by 5 km box I want to find number of points, divide it by total number of points in the entire grid(250km*250km box) and get some number Z and assign relevant color to that box pertaining to the achieved value Z.

Is there any way doing it?

Answer 1

[Walter Roberson]

http://www.mathworks.com/help/stats/hist3.html

[Utsav]

I understand this link. I am sorry for being naive. Please help Sir.

However I am at a loss to understand how to get the color and the axes must remain similar to units of X and Y and not frequency. frequency must be in color.

[Walter Roberson]

Caution: the below will fail if you have any data at X = +125 or greater or Y = +125 or greater. If you have data right at both boundaries then you need extra bins. If your data can be +125 exactly but cannot be -125 exactly then in the below change the 1 + floor() to be 0 + ceil()

binmin = -125; binmax = 125; binwidth = 5;

xbin = 1 + floor((X(:) - binmin) / binwidth);
ybin = 1 + floor((Y(:) - binmin) / binwidth);

nbins = ceil((binmax - binmin) / binwidth);

counts = accumarray([xbin, ybin], 1, [nbins, nbins]);

Z = counts ./ numel(counts);

coords = [binmin, binmin + (nbins-1)*binwidth];

image(coords, coords, Z);
colormap(hot)

[Utsav]

Thanks a lot Sir, for the reply. It worked for me. Also I found the use of "accumarray' in this link

http://blogs.mathworks.com/videos/2010/01/22/advanced-making-a-2d-or-3d-histogram-to-visualize-data-density/