Sequential Estimation in the Agricultural Sciences

• Activity Number: 372
• Type: Topic Contributed
• Date/Time: Wednesday, August 11, 2004 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Madhuri S. Mulekar*+ and Linda J. Young
• Companies: University of South Alabama and University of Florida
• Keywords: coefficient of variation ; ecological model ; Taylor's power model ; Iwao's patchiness regression ; crowding index ; pest population density

Abstract

Pest populations, such as insects and weeds, reduce yield, but control of these pests can be expensive. An understanding of the population dynamics of each species in the field is needed before optimal use of pesticides can occur. One element in the study of population dynamics is precise estimation of the density of pest and beneficial species. Sequential sampling has proven to be a useful tool in this setting. The biological meaning of term "density" is the average number of insects in a sampling unit, which differs from the traditional statistical use of density as a term for the probability density function. The most commonly used measure of precision in sequential estimation of the mean is the coefficient of variation D of the sample mean. Since ease of field implementation is a primary concern in agricultural sciences, often a less than optimal but easy to obtain estimator, is preferred to an optimal estimator requiring computations in the field. Methods commonly used within agriculture to sequentially estimate the mean with a specified coefficient of variation D will be reviewed, and some new approaches will be considered.