|
Activity Number:
|
211
- Contributed Poster Presentations: Business and Economic Statistics Section
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Business and Economic Statistics Section
|
|
Abstract #309695
|
|
|
Title:
|
Decomposing Variance in Wages Using Longitudinal Surveys
|
|
Author(s):
|
Sarah Teichman* and Tyler McCormick
|
|
Companies:
|
University of Washingon and University of Washington
|
|
Keywords:
|
bipartite;
network;
hierarchical ;
heterogeneity ;
Bayesian ;
longitudinal
|
|
Abstract:
|
We introduce a method to model worker and firm interactions and decompose wage variance using data from longitudinal surveys. The subject of wage variance has been studied extensively in countries such as the US and Europe, but most methods rely on comprehensive administrative data. We propose a method that uses employment information collected through pre-existing longitudinal surveys to study wage variance in developing countries. We model worker and firm interactions as a dynamic bipartite network by embedding worker and firm positions in a latent space. This is the basis of our Bayesian hierarchical model, which we use to estimate wages. We apply our model to the Indonesian Family Life Survey (IFLS), and future work involves applying it to several other longitudinal surveys.
|
Authors who are presenting talks have a * after their name.
