Water pollution and income relationships: A seemingly unrelated partially linear analysis.

Pandit, Mahesh; Paudel, Krishna P.

We used a seemingly unrelated partially linear model (SUPLM) to address a potential correlation between pollutants (nitrogen, phosphorous, dissolved oxygen and mercury) in an environmental Kuznets curve study. Simulation studies show that the SUPLM performs well to address potential correlation among pollutants. We find that the relationship between income and pollution follows an inverted U-shaped curve for nitrogen and dissolved oxygen and a cubic shaped curve for mercury. Model specification tests suggest that a SUPLM is better specified compared to a parametric model to study the income-pollution relationship. Results suggest a need to continually assess policy effectiveness of pollution reduction as income increases.

WATER pollution; LINEAR statistical models

Water Resources Research, 2016, Vol 52, Issue 10, p7668

0043-1397

Academic Journal

10.1002/2016WR018655