Separate a matrix into bins based on first column values

I want to seperate a matrix that has two columns of values into bins based on the value of the first column. I would like to be seperate in 1 step incraments from -15 to 15 for the first column and grab the second column value with it so I can analyze the second column values during different sections of column one. Is there a way to do this. I attached the data file that I am trying to seperate into bins. Thank you for your help!

Answer 1

[Voss]

load('DataRoll.mat')
plot(DataRoll(:,1),DataRoll(:,2)) % plot all data for reference
grid on
set(gca(), ...
    'Box','on', ...
    'Layer','top', ...
    'XMinorGrid','on', ...
    'XLim',[-15 16]);

bin_idx = discretize(DataRoll(:,1),-15:16); % using -15:15 misses the data above 15
binned_data = splitapply(@(x){x},DataRoll,findgroups(bin_idx))

binned_data = 31×1 cell array
    {108×2 double}
    {118×2 double}
    {190×2 double}
    {252×2 double}
    {260×2 double}
    {228×2 double}
    {378×2 double}
    {429×2 double}
    {385×2 double}
    {250×2 double}
    {301×2 double}
    {334×2 double}
    {455×2 double}
    {546×2 double}
    {496×2 double}
    {552×2 double}

figure
hold on
cellfun(@(x)plot(x(:,1),x(:,2),'.'),binned_data);
set(gca(), ...
    'Box','on', ...
    'Layer','top', ...
    'XGrid','on', ...
    'XMinorGrid','on', ...
    'MinorGridLineStyle','-', ...
    'MinorGridColor','k', ...
    'MinorGridAlpha',1, ...
    'GridAlpha',1, ...
    'XLim',[-15 16]);

[Austin Bollinger]

Thank you for the help!

[Voss]

You're welcome!

[Austin Bollinger]

Would you know how to find the average of the 2nd column in each bin?

[Voss]

% loading and binning the data as before:
load('DataRoll.mat')
bin_idx = discretize(DataRoll(:,1),-15:16); % using -15:15 misses the data above 15
binned_data = splitapply(@(x){x},DataRoll,findgroups(bin_idx));

% calculating the mean of the second column of data in each bin:
binned_mean = cellfun(@(x)mean(x(:,2),1),binned_data)

binned_mean = 31×1

-3.7843 -4.2714 -5.1898 -8.5879 -8.7605 -8.9249 -8.8929 -7.9495 -8.0889 -9.0207

[Austin Bollinger]

Thank you again!

[Voss]

You're welcome!

Answer 2

[dpb]

load DataRoll
bar(N)
xlim([0.25 31.75])
[min(DataRoll(:,1)) max(DataRoll(:,1))]
[N,e,g]=histcounts(DataRoll(:,1),-15:16);
t=[array2table(DataRoll),table(g)];
groupsummary(t,"g","std","DataRoll2")

You can do whatever you want with the grouping variable calculation -- also see splitapply rowfun and/or varfun as well as groupfilter

NB: the range of your data are from

 [min(DataRoll(:,1)) max(DataRoll(:,1))]
ans =
  -14.9047   15.5137

>> 

so the edges vector needs must cover 16 as upper end to bin all elements in the first column.

[Austin Bollinger]

Thank you!