|
Activity Number:
|
631
- Small Area Estimation: “Producing Estimates for Small Areas from Sampled Data”
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 1, 2019 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #300458
|
Presentation
|
|
Title:
|
Small Area Estimates Using Tree Based Models
|
|
Author(s):
|
Daniell Toth* and Kelly McConville
|
|
Companies:
|
U.S. Bureau of Labor Statistics and Reed College
|
|
Keywords:
|
machine learning;
random forest;
sample design;
indirect estimators;
survey data;
model-assisted
|
|
Abstract:
|
We investigate the use of models based on regression trees to produce estimates of population means for small areas. We discuss different types of domains and methods for handling them using a variety of tree based estimators and compare their performance using the US Bureau of Labor Statistics data.
|
Authors who are presenting talks have a * after their name.
