QBmover function

Confidence intervals for ratios of proportions based on the quasibinomial assumption

Confidence intervals for ratios of proportions with overdispersed binomial data in a one-factor quasibinomial generalized linear model. Intervals are computed using the MOVER-R method on profile deviance intervals (as implemented in mcprofile) for the single proportions.

QBmover(succ, fail, trt, conf.level = 0.95,
 alternative = "two.sided", grid = NULL)

Arguments

• succ: vector of counts of successes
• fail: vector of counts of failures
• trt: factor variable distinguishing the treatment groups
• conf.level: a single numeric value, the confidence level
• alternative: a character string, "two.sided" for two-sided intervals, "less" for upper limits, "greater" for lower limits only
• grid: optional, a numeric vector to be supplied to the profiling used internally in quasibin.ratio to obtain profile deviance intervals for each samples proportion on the logit-scale.

Returns

A data.frame with three columns - est: estimated ratios

• lower: lower confidence limits

• upper: upper confidence limits

References

Donner and Zou (2012): Closed-form confidence intervals for functions of the normal mean and standard deviation. Statistical Methods in Medical Research 21(4):347-359. Gerhard (2014): Simultaneous Small Sample Inference For Linear Combinations Of Generalized Linear Model Parameters. Communications in Statistics - Simulation and Computation. DOI:10.1080/03610918.2014.895836

Author(s)

Frank Schaarschmidt

Note

Experimental

Examples

QBmover(succ=c(0,0,1,  0,6,8), fail=c(20,20,18, 20,14,12), 
 trt=factor(rep(c("A", "B"), c(3,3))), conf.level = 0.95,
 alternative = "two.sided", grid = NULL)