Figures & data
Fig. 1 Simulations satisfying assumptions. Target for mean row is the mean of true prevalences across datasets. The RMSEs of the test-positive rate and HT-known estimator are identical under simple random testing. The variance of the HT-known estimator is higher than that of the HT-estimated due to occasionally very large weights.
![Fig. 1 Simulations satisfying assumptions. Target for mean row is the mean of true prevalences across datasets. The RMSEs of the test-positive rate and HT-known estimator are identical under simple random testing. The variance of the HT-known estimator is higher than that of the HT-estimated due to occasionally very large weights.](/cms/asset/9e45b46b-a712-4181-bafe-f7e9fee226d5/uasa_a_2238943_f0001_c.jpg)
Fig. 2 Daily test counts, including multiple tests per week by the same student. Vertical grid lines correspond to Mondays.
![Fig. 2 Daily test counts, including multiple tests per week by the same student. Vertical grid lines correspond to Mondays.](/cms/asset/c7746c9f-20d2-4d2e-b11d-f00b41ba1866/uasa_a_2238943_f0002_c.jpg)