ARTICLE
ABSTRACT
One of the major problems involved in estimating information about economic activity across small domains is too small sample size and incompleteness of data sources. For instance, the distribution of enterprises by target variables tends to be considerably right-skewed, with high variation, high kurtosis and outliers. Therefore, it is not obvious that the implementation of traditional estimation methods meets the desired requirements, such as being free from bias or having competitive variance. Furthermore, the pressure to produce accurate estimates at a low level of aggregation or needs to substantially reduce sample size have increased the importance of exploring the possibilities of applying new, more sophisticated methods of estimation. The aim of the study was to test the usefulness of winsorization methods to estimate economic statistics from the DG1 survey.
KEYWORDS
domain estimation, business statistics, winsorized estimator.
REFERENCES
CHAMBERS, R., KOKIC, P., SMITH, P., CRUDDAS, M., (2000).Winsorization for Identifying and Treating Outliers in Business Surveys,Proceedings of the Second International Conference on Establishment Surveys (ICES II), 687–696.
COX, B. G., BINDER, A., CHINNAPPA, N. B., CHRISTIANSON, A., COLLEDGE, M. J., KOTT, P. S., (1995). Business Survey Methods, John Wiley & Sons.
GROSS, W. F., BODE, G., TAYLOR, J. M., LLOYD–SMITH, C. W., (1986).Some finite population estimators which reduce the contribution of outliers, [in:] Proceedings of the Pacific Statistical Conference, 20–24 May 1985, Auckland, New Zealand.
KOKIC, P. N., BELL, P. A., (1994). Optimal winsorizing cutoffs for a stratified finite population estimator, Journal of Official Statistics, 10, 419–435.
PRESTON, J., MACKIN, C., (2002). Winsorization for Generalised Regression Estimation, Australian Bureau of Statistics.
PRESTON, J., MACKIN, C., (2002). Winsorization for Generalised Regression Estimation, Paper for the Methodological Advisory Committee, November 2002, Australian Bureau of Statistics
