API reference¶
Allocation¶
Generate a selection order matrix (SOM). |
|
Allocate units to groups with the Finite Selection Model. |
|
Allocate a new batch of units given units already allocated. |
|
Draw a completely randomized design. |
|
Conditional SCOMARS probabilities along a realized two-group path. |
|
The order in which treatment groups take turns choosing units. |
|
Outcome of an FSM allocation. |
Balance and design assessment¶
Target absolute standardized mean difference of one group. |
|
TASMD of one group under one or more designs. |
|
TASMD across repeated assignments. |
|
Absolute standardized mean difference between two groups. |
|
Randomization-based effective sample size of each design. |
|
Model-based effective sample size of each assignment. |
|
Plot imbalance per covariate for one or two assignments. |
Inference¶
Fisher randomization test of a sharp null hypothesis. |
|
Result of a Fisher randomization test. |
Covariates and data¶
Add squares and pairwise products of the columns of |
|
Dehejia and Wahba's subset of the National Supported Work data. |