We study the impact of the odd-even transportation policy (environmental regulation), among other controlled factors on Delhi's pollution levels using panel data. We have daily data on five different pollutants from 9th April 2015 through 29th July 2018 and also pollutant's data over nine different locations of Delhi averaged over different years from 2014 onwards. We consider explanatory variables ranging from the dummy for introduction of the odd-even transportation policy adopted by Delhi Government in 2015 and 2016, along with polynomial terms of time; climatic factors like wind speed, average temperatures, relative humidity and rainfall; fringe factors like ban on sale of crackers, dummy for Deepawali and parallel build-up of highway for movement of trucks at the outskirts of Delhi, burning of agricultural residue in surrounding states, price of fossil fuels, spatial dependence among nine different locations of Delhi, number of CNG/electric cars plying in Delhi, number of registered public and private vehicles, among others. We use spatial regression and parametric RDD (Regression Discontinuity Design) for our study. The RDD exercise would help us to know through the impact of the dummy of the new transportation policy (with and without polynomial terms), its impact on the short- and long-term pollution levels in Delhi. Pollution data is gathered from Central Pollution Control Board (CPCB) offices of Delhi. At the end, we suggest some policy measures which should be undertaken for reducing pollution levels in Delhi. We have used time as the running variable for each pollutant with a treatment date as the threshold.
RDD (Regression Discontinuity Design) Spatial Regression Odd-even Transportation Policy I Moran's statistics Panel data