Documentation |
imp = predictorImportance(ens)
[imp,ma]
= predictorImportance(ens)
imp = predictorImportance(ens) computes estimates of predictor importance for ens by summing these estimates over all weak learners in the ensemble. imp has one element for each input predictor in the data used to train this ensemble. A high value indicates that this predictor is important for ens.
[imp,ma] = predictorImportance(ens) returns a P-by-P matrix with predictive measures of association for P predictors.
imp |
A row vector with the same number of elements as the number of predictors (columns) in ens.X. The entries are the estimates of predictor importance, with 0 representing the smallest possible importance. |
ma |
A P-by-P matrix of predictive measures of association for P predictors. Element ma(I,J) is the predictive measure of association averaged over surrogate splits on predictor J for which predictor I is the optimal split predictor. predictorImportance averages this predictive measure of association over all trees in the ensemble. |
predictorImportance computes estimates of predictor importance for tree by summing changes in the mean squared error (MSE) due to splits on every predictor and dividing the sum by the number of branch nodes. If the tree is grown without surrogate splits, this sum is taken over best splits found at each branch node. If the tree is grown with surrogate splits, this sum is taken over all splits at each branch node including surrogate splits. imp has one element for each input predictor in the data used to train this tree. At each node, MSE is estimated as node error weighted by the node probability. Variable importance associated with this split is computed as the difference between MSE for the parent node and the total MSE for the two children.
The predictive measure of association between the optimal split on variable i and a surrogate split on variable j is:
$${\lambda}_{i,j}=\frac{\text{min}\left({P}_{L},{P}_{R}\right)-\left(1-{P}_{{L}_{i}{L}_{j}}-{P}_{{R}_{i}{R}_{j}}\right)}{\text{min}\left({P}_{L},{P}_{R}\right)}.$$
Here
P_{L} and P_{R} are the node probabilities for the optimal split of node i into Left and Right nodes respectively.
$${P}_{{L}_{i}{L}_{j}}$$ is the probability that both (optimal) node i and (surrogate) node j send an observation to the Left.
$${P}_{{R}_{i}{R}_{j}}$$ is the probability that both (optimal) node i and (surrogate) node j send an observation to the Right.
Clearly, λ_{i,j} lies from –∞ to 1. Variable j is a worthwhile surrogate split for variable i if λ_{i,j} > 0.
Element ma(i,j) is the predictive measure of association averaged over surrogate splits on predictor j for which predictor i is the optimal split predictor. This average is computed by summing positive values of the predictive measure of association over optimal splits on predictor i and surrogate splits on predictor j and dividing by the total number of optimal splits on predictor i, including splits for which the predictive measure of association between predictors i and j is negative.
Estimate the predictor importance for all numeric variables in the carsmall data:
load carsmall X = [Acceleration Cylinders Displacement ... Horsepower Model_Year Weight]; ens = fitensemble(X,MPG,'LSBoost',100,'Tree'); imp = predictorImportance(ens) imp = 0.0082 0 0.0049 0.0133 0.0142 0.1737
The weight (last predictor) has the most impact on mileage (MPG). The second predictor has importance 0; this means the number of cylinders has no impact on predictions made with ens.
Estimate the predictor importance for all variables in the carsmall data for an ensemble where the trees contain surrogate splits:
load carsmall surrtree = templateTree('Surrogate','on'); X = [Acceleration Cylinders Displacement ... Horsepower Model_Year Weight]; ens2 = fitensemble(X,MPG,'LSBoost',100,surrtree); [imp2,ma] = predictorImportance(ens2) imp2 = 0.0725 0.1342 0.1425 0.1397 0.1380 0.1855 ma = 1.0000 0.0414 0.0607 0.0782 0.0102 0.0322 0 1.0000 0 0 0 0 0.0441 0.0704 1.0000 0.0883 0.0175 0.0913 0.0944 0.1166 0.1400 1.0000 0.0390 0.1308 0.0121 0.0139 0.0127 0.0127 1.0000 0.0113 0.0818 0.1317 0.2072 0.1878 0.0340 1.0000
While weight (last predictor) still has the most impact on mileage (MPG), this estimate has the second predictor (number of cylinders) is essentially tied for third most important predictor.