Activity Number:
|
372
- Government Statistics Section P.M. Roundtable Discussion (Added Fee)
|
Type:
|
Roundtables
|
Date/Time:
|
Tuesday, July 31, 2018 : 12:30 PM to 1:50 PM
|
Sponsor:
|
Government Statistics Section
|
Abstract #329451
|
|
Title:
|
Using the R Package: Recursive Partitioning for Modeling Survey Data (RPMS)
|
Author(s):
|
Daniell Toth*
|
Companies:
|
Bureau of Labor Statistics
|
Keywords:
|
complex sample;
sample design;
regreesion trees;
tree models
|
Abstract:
|
In this roundtable we will discuss the R-package rpms. The package allows the user to fit a linear model to survey data in each node obtained by recursively partitioning the data. The splitting variables and splits selected are obtained using a procedure which accounts for complex sample design features used to obtain the data. The main function returns the resulting binary tree with the linear model fit at every end-node. In addition the package provides a number of functions and methods for analyzing survey data using the obtained tree model. Discussion will center on using the package for performing analysis of survey data, including model-building and interpretation.
|
Authors who are presenting talks have a * after their name.