1 Accumarray Examples

Go to the MLX, M, PDF, or HTML version of this file. Go back to fan’s MEconTools Package, Matlab Code Examples Repository (bookdown site), or Math for Econ with Matlab Repository (bookdown site).

1.1 Accumarry Basic Example

There are three unique values in ar_a, sum up the probabilities for each of the unique states. This is equivalent to sorting a matrix with a and prob, and computing sum for each.

ar_a = [3,2,1,3]';
ar_prob = [0.1,0.2,0.31,0.39]';
ar_sumprob = accumarray(ar_a, ar_prob);
tb_summed_prob = table(sort(unique(ar_a)), ar_sumprob);
disp(tb_summed_prob);

    Var1    ar_sumprob
    ____    __________

     1         0.31   
     2          0.2   
     3         0.49

1.2 Accumarry For Discrete Random Variable

Upon solving a model, if we look for the mass at certain choices or states, accumarray could help aggregate up probabilities

a1 = [1,1,2,2]

a1 = 1x4    
     1     1     2     2

a2 = [3,2,1,3]

a2 = 1x4    
     3     2     1     3

a3 = [1,2,3,3]

a3 = 1x4    
     1     2     3     3


a = [a1;a2;a3]'/2

a = 4x3    
    0.5000    1.5000    0.5000
    0.5000    1.0000    1.0000
    1.0000    0.5000    1.5000
    1.0000    1.5000    1.5000


prob_a = zeros(size(a)) + 1/12

prob_a = 4x3    
    0.0833    0.0833    0.0833
    0.0833    0.0833    0.0833
    0.0833    0.0833    0.0833
    0.0833    0.0833    0.0833

[ar_idx_full, ~, ar_idx_of_unique] = unique(a)

ar_idx_full = 3x1    
    0.5000
    1.0000
    1.5000

ar_idx_of_unique = 12x1    
     1
     1
     2
     2
     3
     2
     1
     3
     1
     2

mt_idx_of_unique = reshape(ar_idx_of_unique, size(a))

mt_idx_of_unique = 4x3    
     1     3     1
     1     2     2
     2     1     3
     2     3     3


accumarray(mt_idx_of_unique(:,1), prob_a(:,1))

ans = 2x1    
    0.1667
    0.1667

accumarray(mt_idx_of_unique(:,2), prob_a(:,2))

ans = 3x1    
    0.0833
    0.0833
    0.1667

accumarray(mt_idx_of_unique(:,3), prob_a(:,3))

ans = 3x1    
    0.0833
    0.0833
    0.1667