الفهرس 
ON SOME ESTIMATION PROBLEMS OF EXPONENTIATED BURR TYPE XII DISTRIBUTION
ContentsOnly 14 pages are availabe for public view
 |  | from | 112 |  |  |
| | | from | 112 | | |
Abstract
In most statistical studies, the population parameters are unknown and must be estimated from a sample. One of the traditional methods of sampling is simple random sample (SRS) which enables the experimenter to make valid judgment of the parameter of interest. Other than (SRS) more structured techniques in sampling designs such as stratified random sampling and clustered random sampling are also available to make sure that the obtained data provide a good representation of the population of interest. McIntyre (1952) first introduced Ranked set sampling (RSS) as an efficient alternative to SRS that can be used when taking exact measurement of the units of interest is very costly while ranking small sets of these selected units is cheap. he noted that, in many sampling situations ,there is enough information available to enable sampling units to be ranked with a reasonable accuracy according to the character of interest. The ranking may be done on the basis of visual inspection, prior information, expert opinion or some combination of such methods. Hence the Ranked set sampling (RSS) is a statistical technique for data collection that generally leads to more efficient estimators than those based on simple random sampling (SRS).The Exponentiated Burr XII (EBXII) is an extended distribution to Burr type XII distribution. This model is important model for several areas such as actuarial sciences, economic and survival analysis so, the aim of this thesis is to estimate the parameters of exponentiated Burr XII distribution by the methods of ranked set sampling (RSS) and compare these estimates with those of simple random sample (SRS). Fisher information matrix for both (SRS) and (RSS) for the unknown parameters of (EBXII) are also derived. Simulation study compared between the estimators of both methods in terms of their biases, mean square errors and efficiencies. It is shown that the estimators based on RSS are more efficient than those of SRS