Table 8 Time performance of the Horowitz–Manski (HM) and Manski–Tamer (MT) interval estimators for the linear peer effects models (with non-closed groups and individuals excluded from their own peer group). 50 Monte Carlo simulations
From: Partial identification of nonlinear peer effects models with missing data
Group size | No. of groups | 5 missing observations | ||
|---|---|---|---|---|
Prob. of missing | Average time (secs) | |||
(in %) | OLS | |||
HM | MT | |||
5 | 50 | 2.00 | 6.8 | 0.3 |
5 | 100 | 1.00 | 6.9 | 0.4 |
5 | 200 | 0.50 | 6.9 | 0.4 |
5 | 500 | 0.20 | 8.2 | 0.6 |
10 | 50 | 1.00 | 6.7 | 0.4 |
10 | 100 | 0.50 | 6.5 | 0.4 |
10 | 200 | 0.25 | 7.0 | 0.6 |
10 | 500 | 0.10 | 9.1 | 0.8 |
25 | 50 | 0.40 | 6.4 | 0.4 |
25 | 100 | 0.20 | 6.9 | 0.5 |
25 | 200 | 0.10 | 8.2 | 0.8 |
25 | 500 | 0.04 | 12.6 | 1.7 |