Confidence Intervals
•
This is the observed interval which
potentially
includes the
unobservable true parameter of interest
•
A CI of 95% is NOT the probability that the interval contains
the true parameter
➢
Observed data points are random samples of true population – so is
the CI
•
A CI of 95% refers to a number of experiments each of which
has a CI set then the proportion of those intervals which
contain the true value approaches the confidence interval
➢
ie how frequently the observed intervals contain the true value of the
parameter
➢
Or a CL of 95% means 95% of the observed CI’s will hold the true
value of the parameter