برآورد پارامترهاي توزيع نرمال براساس مشكلات ركوردي

Estimation of Normal Parameters on the Basis of Record Observations

Let X1,X2,... be a sequence of independended and identically distributed random variables, with distribution function F. Xj is called an high(low) record, when it is bigger(smaller) than itʹs privious observations. In this article, we assume that F belongs to a location-scale family of distributions. On the basis of record observations, we obtain some estimators for the parameters of this family. For obtaining maximum likelihood estimators, we need to solve a nonlinear system of equation. In this case, we cosider normal case, and by using numerical analysis, we calculate maximum likelihood estimators for the parameter of normal distribution. Also, by using minimum mean square criterion, we obtain the best linear unbiased estimators for the parameters of normal distribution, based on record observations.