Application of air quality combination forecasting to Bogota

• Published in: Atmospheric environment (1994) Vol. 89; pp. 22 - 28
• Main Authors: Westerlund, Joakim, Urbain, Jean-Pierre, Bonilla, Jorge
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.06.2014; Elsevier
• Subjects: air quality; Air quality forecasting; Analysis methods; Applied sciences; atmospheric chemistry; Atmospheric pollution; Bogota; cities; Economics and Business; Ekonomi och näringsliv; Environmental Sciences; Exact sciences and technology; Forecast combination; GREECE; IMPUTATION; Latin America; Meteorology & Atmospheric Sciences; MISSING VALUES; Monte Carlo method; Neural networks; POLLUTION; regression analysis; SO2; South America; Colombia; America; Latin America; C53; Air quality forecasting; Neural networks; C45; Forecast combination; Bogota; Q53; Linear regression; Air quality; Urban area; Forecast model; Statistical forecasting; Statistical method; Atmospheric pollution forecasting; Air pollution
• ISSN: 1352-2310; 1873-2844
• DOI: 10.1016/j.atmosenv.2014.02.015

The bulk of existing work on the statistical forecasting of air quality is based on either neural networks or linear regressions, which are both subject to important drawbacks. In particular, while neural networks are complicated and prone to in-sample overfitting, linear regressions are highly dependent on the specification of the regression function. The present paper shows how combining linear regression forecasts can be used to circumvent all of these problems. The usefulness of the proposed combination approach is verified using both Monte Carlo simulation and an extensive application to air quality in Bogota, one of the largest and most polluted cities in Latin America. •The paper considers a relatively novel forecasting method, combination forecasting.•Combination forecasting is shown to outperform the standard neural network approach.•Combination forecasting is applied to a unique data set for Bogota.