Crop Acreage Prediction Combining Several Sources of Information

• Activity Number: 624
• Type: Invited
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Jae-kwang Kim and Zhonglei Wang*
• Companies: Iowa State University and Iowa State University
• Keywords: Measurement error models ; Hierarchical model ; Small area estimation ; Big data

Abstract

Combining information from different source is an important practical problem. The source of information can come from a probability sampling with direct measurement, from another probability sampling with indirect measurement (such as self-reported health status), from auxiliary area level information, or from a non-probability sample (such as administrative records from a particular program). Such problems has been addressed in the context of area level small area estimation. In this talk, we are interested in combining information from three sources. The first source is the June Area Survey (JAS). In JAS, the sample observations are obtained from a probability sampling. The second source is from the Farm Service Agency (FSA) data. The FSA data is obtained from a voluntary participation of certain programs. The third source is from the classification of the Cropland Data Layer (CDL).

To combine several source of information, we use area level model approach that is based on the statistical modeling of area level estimates. Specifically, we use hierarchical measurement error model approach to combine the information and borrow strength from observations outside the stratum.