This paper deals with the problem of interval estimation of a    model, where  and  are non-negative independent, continuous random variables following either a general exponential form distribution or a general inverse exponential form distribution. Simple confidence limits for  are obtained by applying different procedures. The results obtained are illustrated using the Weibull distribution as an example for the general exponential form, whereas the inverse Weibull distribution as an example for the general inverse exponential form. The average length and the probability coverage of the confidence intervals are examined by simulation.

approximate confidence interval, generalized variable approach, maximum likelihood estimator, percentile bootstrap, t-bootstrap.