how to sort a dataset into different group and get sum of each group?

Hi everyone,

Now I have a question on how to sort a dataset into different group and get sum of each group, which tortures me for sevral day. I hope to get help from you, and thanks.

An example of my dataset is shown below. There are 9 rows and 12 columns. This first and second rows are x, y coordinates and the third row is the correponding value of each (x ,y).

-1.986  -2.648  -3.310  -3.972  -4.634  -5.296  -5.958  -6.621  -7.283  -7.945  -8.607  -9.269  -9.931  -10.593  -11.255  -11.917  -12.579//
-4.515  -6.020  -7.525  -9.030  -10.536  -12.041  -13.546  -15.051  -16.556  -18.061  -19.566  -21.071  -22.576  -24.081  -25.586  -27.091//  
0.002  0.003  0.003  0.004  0.004  0.005  0.005  0.005  0.006  0.006  0.006  0.007  0.007  0.007  0.007  0.008  0.008  0.008

-3.967  -5.289  -6.611  -7.934  -9.256  -10.578  -11.900  -13.223  -14.545  -15.867  -17.189  -18.512  -19.834  -21.156  -22.478  -23.801//
-5.372  -7.163  -8.954  -10.745  -12.536  -14.326  -16.117  -17.908  -19.699  -21.490  -23.280  -25.071  -26.862  -28.653  -30.443  -32.234//
0.001  0.002  0.002  0.003  0.003  0.003  0.004  0.004  0.004  0.005  0.005  0.005  0.005  0.005  0.005  0.006

-3.606  -4.808  -6.011  -7.213  -8.415  -9.617  -10.819  -12.021  -13.223  -14.425  -15.627  -16.829  -18.032  -19.234  -20.436  -21.638//
-2.758  -3.678  -4.597  -5.517  -6.436  -7.356  -8.275  -9.195  -10.114  -11.034  -11.953  -12.873  -13.792  -14.711  -15.631  -16.550//
0.002  0.003  0.003  0.004  0.005  0.005  0.006  0.006  0.006  0.007  0.007  0.007  0.008  0.008  0.008  0.008

I want to sort the data according to (x,y) coordinates into idfferent groups and get the sums of each group. In this case, the group can be

(-35<x<-30,-35<y<-30),(-30<x<-25,-35<y<-30),....(-5<x<0,-35<y<-30);
(-35<x<-30,-30<y<-25),(-30<x<-25,-30<y<-25),....(-5<x<0,-30<y<-25);
(-35<x<-30,-25<y<-20),(-30<x<-25,-25<y<-20),....(-5<x<0,-25<y<-20);
(-35<x<-30,-5<y< 0),(-30<x<-25,-5<y< 0),....(-5<x<0,-5<y<0)

Then get the sum of each group.

Thanks once again.

[Walter Roberson]

Is it correct that you are wanting to distribute into a rectangular grid? If so it becomes easy. Even easier if the grid boundaries are the same distance apart.

What should be done for the case that x is exactly 30? As you have used inequalities at each stage, points on the boundaries have no-where to go.

[Genhou]

Yes.I want to distribute into rectangular grid.

The data and the rectangular grid that I show are an examples to merely explain my question. Your suggestion is great. I would choose this way:(-35=<x<-30,-5=<y< 0),(-30=<x<-25,-5=<y=< 0),....(-5=<x<0,-5<y=<0).

Then How to get the distribution into rectangular grid?

Answer 1

[Walter Roberson]

mingrid = -30;
gridgap = 5;

xgrididx = 1 + floor((x - mingrid) ./ gridgap);
ygrididx = 1 + floor((y - mingrid) ./ gridgap);

gridcounts = accumarray( [xgrididx(:), ygrididx(:)], 1 );

Now gridcounts is an array of counts.

row K is the range (K - 1) * gridgap + mingrid <= x < K * gridgap + mingrid . Same calculation for column number and y.