Michael Price
Michael is working with the National Agricultural Statistics Service to help develop estimators that will be used in modeling crop acres planted in the United States. Specifically, he is working with Farm Service Administration (FSA) data. Farmers report the number of acres they planted to the FSA for insurance purposes, however not all farmers do this. In order to develop an estimator from this FSA data, the report rate needs to be estimated.
Zhonglei Wang
Zhonglei’s research focuses on the prediction of the acreage of different crops at the level of districts (groups of nearby counties in a state). There are three sources of data available: June Area Survey, Farm Service Agency and Cropland Data Layer. A two-parametric structure model characterizes the relationship between districts, and a generalized least square is used to get the final prediction. The method of moment has been used for the estimation of the parameters in the structural models, and now an EM algorithm using parametric fractional imputation is under consideration.