vankesteren / ProportionalFitting.jl

Multidimensional iterative proportional fitting in Julia

ProportionalFitting.jl

Multidimensional iterative proportional fitting in Julia.

ProportionalFitting implements a multidimensional version of the factor estimation method for performing iterative proportional fitting (also called RAS algorithm, raking, matrix scaling)

Before version 0.3.0, this package was called ItPropFit.jl.

Showcase

See the full documentation and getting started here.

using ProportionalFitting

# matrix to be adjusted
X = [40 30 20 10; 35 50 100 75; 30 80 70 120; 20 30 40 50]

# target margins
u = [150, 300, 400, 150]
v = [200, 300, 400, 100]

# Perform iterative proportional fitting
fac = ipf(X, [u, v])

[ Info: Converged in 8 iterations.
Factors for 2D array:
  [1]: [0.9986403503185242, 0.8833622306385376, 1.1698911437112522, 0.8895042701910321]
  [2]: [1.616160156063788, 1.5431801747375655, 1.771623700829941, 0.38299396265192226]

# compute adjusted matrix
Z = Array(fac) .* X

4×4 Matrix{Float64}:
 64.5585   46.2325   35.3843   3.82473
 49.9679   68.1594  156.499   25.3742
 56.7219  144.428   145.082   53.7673
 28.7516   41.18     63.0347  17.0337

# check that the margins are indeed [u, v]
ArrayMargins(Z)

Margins of 2D array:
  [1]: [150.0000000009452, 299.99999999962523, 399.99999999949796, 149.99999999993148]
  [2]: [200.0, 299.99999999999994, 399.99999999999994, 99.99999999999997]