Extending Hilbert–Schmidt Independence Criterion for Testing Conditional Independence

. 2023 Feb 26;25(3):425. doi: 10.3390/e25030425

Algorithm 1:Local bootstrap

Input: Data $(x_{i}, y_{i}, z_{i})$ , $i = 1, \dots, n$ .

Output: New samples: $(x_{i}^{*}, y_{i}^{*}, z_{i}^{*}), i = 1, \dots, n$

1
for $i \leftarrow 1$ to n do
2
Let $z_{i}^{*} = z_{i}$
3
Sample $x_{i}^{*} \sim {\hat{G}}_{x ∣ z_{i}^{*}}, y_{i}^{*} \sim {\hat{G}}_{y ∣ z_{i}^{*}}$ .
4
end