The output indicates that the p-values are above 0.15 in all windows between the minimum window [-1.000, 1.000] and the window [-1.090, 1.090]. The p-value is correct, but I am confused about the observations (136 below, 45 above) for the recommended window. Because X1 is my running variable, which is mentioned below. I think it should be 75 below and 46 above observation in the recommended window, or are my results correct?
rdwinselect X1 $covariates1, cutoff(0.000005) wmin(1) wstep(0.01) seed(50) level(0.001)
Window selection for RD under local randomization
Recommended window is [-1.090; 1.090] with 181 observations (136 below, 45 above).
rdwinselect X1 $covariates1, cutoff(0.000005) wmin(1) wstep(0.01) seed(50) level(0.001)
Window selection for RD under local randomization
| Cutoff c = 0.00 | Left of c | Right of c | Number of observation | = | 3400 |
| Number of obs | 3286 | 114 | Kernel | type | = | uniform |
| 1st percentile | 0 | 0 | Reps | = | 1000 |
| 5th percentile | 286 | 0 | Testing | method | = | rdrandinf |
| 10th percentile | 385 | 0 | Balance | test | = | diffmeans |
| 20th percentile 513 | 0 |
| Bal. test p-value | Var.name | Bin.test p-value | Obs<c | Obs>=c |
| Window |
| -1.000 | 1.000 | 0.160 | WEP | 0.013 | 73 | 45 |
| -1.010 | 1.010 | 0.388 | WIP | 0.000 | 136 | 45 |
| -1.020 | 1.020 | 0.412 | WIP | 0.000 | 136 | 45 |
| -1.030 | 1.030 | 0.420 | WIP | 0.000 | 136 | 45 |
| -1.040 | 1.040 | 0.380 | WIP | 0.000 | 136 | 45 |
| -1.050 | 1.050 | 0.466 | WIP | 0.000 | 136 | 45 |
| -1.060 | 1.060 | 0.414 | WIP | 0.000 | 136 | 45 |
| -1.070 | 1.070 | 0.386 | WIP | 0.000 | 136 | 45 |
| -1.080 | 1.080 | 0.408 | WIP | 0.000 | 136 | 45 |
| -1.090 | 1.090 | 0.386 | WIP | 0.000 | 136 | 45 |
Comment