| DATE: | Friday, April 29, 2005 |
| TIME: | 4:40 - 5:40 PM |
| PLACE: | LN 2205 |
| SPEAKER: | Nitis Mukhopadhyay (University of Connecticut) |
| TITLE: | A New Two-Stage Sampling Design for Estimating the Maximum Average Time to Flower |
The point of this talk is to emphasize that a practical problem can lead to non-trivial and new statistical methodology along with sophisticated theory at the same time. The new "stuff" can then be fruitfully applied in the "field" to reach a practical solution.
A horticulturist was considering the number of days (X) each variety took from planting seeds to reach a stage when the first bud appeared for three local marigold varieties. The data X could be recorded within one-half day. The primary interest was to estimate the maximum waiting time between "seeding" and "first budding" among the varieties under consideration. The available information guided us to use unequal pilot sample sizes.
We were not aware of any existing methodology with unequal pilot sample sizes that would readily apply here. A new two-stage sampling design had to be developed and implemented. Important exact as well as large-sample properties of the proposed methodology are highlighted. This methodology is proven to be theoretically superior to the existing methodology for large sample sizes if the pilot sizes "could be chosen" equal. With the data obtained from our designed experiment, the superiority of the new methodology will be indicated.
This material is based on some joint work with Professors Mun S. Son and Y. C. Ko from the University of Vermont, Burlington.