Normalizing a polar histogram

Hello,

I have two groups with distributions of phase angles which I want to plot in two seperate polar histograms, using the matlab function 'polarhistogram' (2017a). Moreover, I want to normalize the histograms by the number of observations across both groups. Normalizing each histogram by the number of observations in each group seperately is straightforward by inputting 'normalization', 'probability'. But although it sounds simple, I have found no way yet of normalizing each polar plot by the total number of observations.

For normal histograms I know a trick for this (there one can use the output of 'hist' first, change it, and input it into 'bar'), however this does not work for this function.

I would be happy if anyone can suggest a solution for it.

Answer 1

[Steven Lord]

I haven't tried this, but if I understand what you're trying to do I believe it will work.

1. Bin the data yourself (using the same set of edges) using two calls to histcounts.
2. Normalize each set of bins using the combined data from both sets of bins.
3. Specify the 'BinEdges' and 'BinCounts' inputs in two calls to polarhistogram with a different polaraxes for each, one call per set of data.

[Denis Wolff]

Thanks very much, it works :-)