Rounded Data From EV Model

In real problems, observations are often rounded when recorded due to the precision of the recording mechanism. We might called such data "rounded data" since they are obtained by measuring to the nearest unit. It is no doubt that the rounded data has impact on the estimation of parameters in statistical inferences. This paper deals with the rounded data in EV model to present the estimation of parameters. The advantage of the new method used in this paper is applied widely. In EV model, the new method in contrast with traditional methods can be used in different circumstances. In this paper, we discuss three methods to deal with the influence of rounded data: Sheppard correction, BRB correction and maximum likelihood estimation of rounded data. MLE will be the most important. For linear EV model, we can obtain likelihood function of rounded data and corresponding MLE.