dpuelz / BicliqueRT

Randomization Tests of Causal Effects under General Interference
https://arxiv.org/pdf/1910.10862.pdf
6 stars 4 forks source link

Some updates #18

Closed szhuang1 closed 2 years ago

szhuang1 commented 2 years ago

In gen_tstat, we allow functions to return NaN when the clique conditional on is ill-behaved. In biclique.decompose, maintain the (matrix) shape of multiNEgraph and Z_m when slicing the matrix. Update the Clustered Interference example in README that out_bassefeller should use exposure under Z not Z itself.