Skip to main content

Table 2 Performance evaluation criteria

From: Left-censored recurrent event analysis in epidemiological studies: a proposal for when the number of previous episodes is unknown

Simulation study 1

Evaluation criteria (covariate j = 1, 2, 3 and sample m = 1, …, 100)

Mean relative bias

\(\frac{\sum_{j=1}^3\sum_{m=1}^{100}\frac{\left|{\hat{\beta}}_{jm}-{\beta}_j\right|}{\beta_j}\times 100}{3\times 100}\)

Average length of 95%CI

\(\frac{\sum_{j=1}^3\sum_{m=1}^{100}2{Z}_{1-\alpha /2} SE\left({\hat{\beta}}_{jm}\right)}{3\times 100}\)

Coverage

Percentage of times the 95% confidence interval \({\hat{\beta}}_{jm}\pm {Z}_{1-\alpha /2} SE\left({\hat{\beta}}_{jm}\right)\) includes βj

Simulation study 2

Evaluation criterium

Type I error

Percentage of times the H0:β = 0 is rejected, for m = 1, …, 1000