Parameter summaries Provides the summary estimates of the alternative model for bivariate random-effects meta-analysis by Riley et al. (2008) with their corresponding standard errors (derived from the inverse Hessian). For confidence intervals, asymptotic normality is assumed.

Parameter summaries Provides the summary estimates of the alternative model for bivariate random-effects meta-analysis by Riley et al. (2008) with their corresponding standard errors (derived from the inverse Hessian). For confidence intervals, asymptotic normality is assumed.

Usage

# S3 method for riley
summary(object, ...)

Arguments

object

A riley object

...

Arguments to be passed on to other functions (currently ignored)

Value

array with confidence intervals for the estimated model parameters. For diagnostic test accuracy data, the resulting summary sensitivity and false positive rate are included.

Details

For meta-analysis of diagnostic test accuracy data, beta1 equals the logit sensitivity (Sens) and beta2 equals the logit false positive rate (FPR).

Note

For the overall correlation (rho) confidence intervals are derived using the transformation logit((rho+1)/2). Similarly, the logit transformation is used to derive confidence intervals for the summary sensitivity and false positive rate.

References

Riley RD, Thompson JR, Abrams KR. An alternative model for bivariate random-effects meta-analysis when the within-study correlations are unknown. Biostatistics 2008; 9: 172--186.

Author

Thomas Debray <thomas.debray@gmail.com>