 |
 |
Dr Nicholas Longford of Universitat Pompeu Fabra, Barcelona and SNTL Consulting will give a one-day course on Small Area Estimation at the Department of Statistics, University of Waikato, Hamilton, New Zealand on 28 August 2009. (This is the Friday preceding the NZ Statistical Association Conference in Wellington which runs from 2-3 September, with a workshop on Tuesday 1 September 2009.) The lectures will be based on Nick Longford's book "Missing Data and Small-Area Estimation (Springer, 2005; Part II)" Nick has outlined his intentions for the one-day course as follows: Small-area estimation is concerned with inferences about districts or other divisions (small areas) of a country in a national survey. Such surveys rarely have resources for large enough samples in every district for the analysis of these subsamples in isolation from the rest of the survey data. Small-area methods draw on the information from the entire sample for estimating quantities related to each district. The key concept in this is borrowing strength across the districts (or exploiting their similarity). In its full generality, it includes using information from other surveys and censuses, as well as past editions of the same survey programme. The course will discuss the basics of small-area estimation, including a historical overview, the relevance of empirical Bayes models and the interplay of the sampling-design and model-based perspectives. Several examples of the real-life kind will be discussed. The issues of designing surveys for small-area estimation will be addressed and some advanced topics (incl. spatial similarity) addressed. There are no absolutes for the course participants, but familiarity with the basics of sampling theory and ordinary regression, and some experience with statistical computing (in anything between Excel and R) would be a distinct advantage. Participants are encouraged to submit specific problems or other matters for discussion (a day or two) in advance. Given interest, an additional session (a clinic) could be organized in which these would be addressed in greater detail than the course might permit. Details of timing and registration fees are still to be finalised but we expect to begin at 10am on Friday 28 August and conclude around 4:30. There will be a social and dinner after the event. Please send expressions of interest to the Department Administrator Tania Robinson <[log in to unmask]> with a copy to Dr Judi McWhirter <[log in to unmask]>. Further details will be available shortly. Murray Jorgensen -- Dr Murray Jorgensen http://www.stats.waikato.ac.nz/Staff/maj.html Department of Statistics, University of Waikato, Hamilton, New Zealand Email: [log in to unmask] Fax 7 838 4155 Phone +64 7 838 4773 wk Home +64 7 825 0441 Mobile 021 0200 8350