Abstract
In estimation of the parameter of an exponential distribution, even if no exact data value may be obtained, the interval which contains the true value may be specified. If the number of occurrences of event are counted at regular intervals, the maximum likelihood estimator can be obtained. In this paper, asymptotic inference method for the parameter is presented, and some statistical characteristics of this estimator are simulated in various data collecting conditions. It is seen that interval-data give almost the same estimate as those from complete data.
