SOC222H5 Lecture 4: Lecture 4 – Estimation
Document Summary
We use estimation to make choices based on incomplete information about the population of interest (i. e. we must usually use sample data). Point estimates and confidence intervals, of the population parameter, are both based on the statistic(s) obtained from the sample. Two of the key criteria by which the choice of sample statistics (estimators) is assessed are: bias and efficiency. An estimator is unbiased if, and only if, the mean of its sampling distribution is equal to the population value of interest. The sample mean and sample proportion are known to be unbiased estimators; all other statistics, including the sample standard deviation, are biased, to varying degrees. Unbiased estimator cannot be guaranteed because of random chnace. The extent to which the sampling distribution is clustered about its mean an issue of dispersion. Since the standard error is related to the sample size, the efficiency of an estimator can be increased by increasing the sample size.
