2D Kernel Density Estimate in Matlab

问题:

I am using this function to estimate kernel density in 2D. I am slightly confused by the parameters of this function however.

Here is an example, viewed from directly above, where density is being calculated at each point (O) in the figure. i.e: over very small areas.

I want to change the KDE function parameters so that density is computed over a larger area (for example, the area circled in red). Which parameters do I need to change? I presume it is one (or both) of these:

"n: size of the n by n grid over which the density is computed (default 2^8)"

OR:

"MIN_XY, MAX_XY: limits of the bounding box over which the density is computed". The default limits are computed as:

MAX = max(data,[],1);  MIN = min(data,[],1);  Range = MAX-MIN;  MAX_XY = MAX+Range/4;  MIN_XY = MIN-Range/4; 

Thanks very much.

回答1:

I have run some tests with this function and the solution is to use lower values of n. Here is a series of comparison figures, using the same dataset. The value of n is shown in the title (all other parameters are kept constant):

文章来源: 2D Kernel Density Estimate in Matlab