Algorithm 1.

Estimating ψ_0,s

1: Choose a technique to estimate the conditional means μ0 and μ0,s, eg, ensemble learning with various predictive modeling algorithms (Wolpert, 1992);

2: $\widehat{\mu }$ ← Regress Y on X using the technique from step (1) to estimate μ0;

3: ${\widehat{\mu }}_s$ ← Regress $\widehat{\mu }(X)$ on X−s using the technique from step (1) to estimate μ0,s;

4: ${\widehat{\psi }}_{n,S}\leftarrow \frac{\frac{1}{n}{\sum }_{i=1}^n{\left\{Y_i-{\widehat{\mu }}_s\left(X_i\right)\right\}}^2-\frac{1}{n}{\sum }_{i=1}^n{\left\{Y_i-\widehat{\mu }\left(X_i\right)\right\}}^2}{\frac{1}{n}{\sum }_{i=1}^n{\left(Y_i-{\overline{Y}}_n\right)}^2}$, as in Equation (10).