ABSTRACT

Statistical inference can be divided into two closely related types of problems; the estimation of the unknown parameters of the mathematical model and the testing of hypotheses about the mathematical model. The method of estimation gives estimates which, besides being consistent, have the valuable property that, for large, they are the most efficient. However, the estimates may be biased as may those obtained by the method of moments. An estimate of a population parameter expressed by a single number is called a point estimate. However, a point estimate gives no idea of the precision of the estimate. One desirable property for an estimator is that of unbiasedness. An estimator is said to be unbiased or accurate if the mean of its sampling distribution is equal to the unknown parameter. The desirable property of a good estimator is that of consistency; which, roughly speaking, says that the larger the sample size the closer the statistic will be to the true value.