What is it about?
Pooled time series pose different problems from standard panel data; first of all nonstationarity issues, secondly how to control for the influence of common factors (world interest rates, commodity prices, common shocks like 9/11 etc.). nonstationarity can be addressed through cointegration techniques. The CCE method of Pesaran allows consistent estimation and nonstationarity testing under common factors. My paper is about performing - and documenting - such analysis with open source R functions. It demonstrates usage by fully replicating the work of Holly, Pesaran and Yamagata on house prices in the US.
Featured Image
Why is it important?
There is no other user-friendly and publicly available software for performing all steps of common-factor robust estimation of cointegrated systems, including spatial panel estimation of defactored residuals. All functionality illustrated here is seamlessly integrated in the 'plm' package for panel data econometrics, and hence can easily interoperate with the 'splm' package for spatial panels. Moreover, this paper makes it easy to replicate HPY's terse paper, which is such a great example of panel time series techniques.
Perspectives
Read the Original
This page is a summary of: Narrow Replication of ‘A Spatio-Temporal Model of House Prices in the Usa’ Using R, Journal of Applied Econometrics, October 2014, Wiley,
DOI: 10.1002/jae.2424.
You can read the full text:
Contributors
The following have contributed to this page