A simple, flexible distributed lag technique: The polynomial inverse lag

This paper presents a new distributed lag technique, the polynomial inverse lag, which has two useful characteristics. It has a flexible shape, allowing both humped and monotonically declining lag weight distributions; and it can be easily implemented with a small number of nested OLS regressions. T...

Full description

Saved in:
Bibliographic Details
Published in:Journal of econometrics 1986-04, Vol.31 (3), p.329-340
Main Authors: Mitchell, Douglas W., Speaker, Paul J.
Format: Article
Language:English
Subjects:
Applications
Econometrics
Estimating techniques
Exact sciences and technology
Mathematical models
Mathematics
Probability and statistics
Regression analysis
Sciences and techniques of general use
Statistics
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a new distributed lag technique, the polynomial inverse lag, which has two useful characteristics. It has a flexible shape, allowing both humped and monotonically declining lag weight distributions; and it can be easily implemented with a small number of nested OLS regressions. The lag is similar in spirit to the Almon lag, but it is an infinite lag and thus does not require specification of a fixed lag length. Experiments with simulated data show a good ability to replicate the true lag structure even in the face of lag structure misspecifition. The technique is also demonstrated using the well-known St. Louis equation.
ISSN:0304-4076
1872-6895
DOI:10.1016/0304-4076(86)90064-3