powered by
Computation of a 1-alpha FAB t-interval using z-optimal spending function
fabtzCI(y, s, dof, alpha = 0.05, psi = list(mu = 0, tau2 = 1e+05, sigma2 = 1))
a numeric scalar, a normally distributed statistic
a numeric scalar, the standard error of y
positive integer, degrees of freedom for s
the type I error rate, so 1-alpha is the coverage rate
a list of parameters for the spending function, including
mu, the prior expectation of E[y]
tau2, the prior variance of E[y]
sigma2 the variance of y
# NOT RUN { n<-10 y<-rnorm(n) fabtzCI(mean(y),sqrt(var(y)/n),n-1) t.test(y)$conf.int # }
Run the code above in your browser using DataLab