64th ISI World Statistics Congress - Ottawa, Canada

64th ISI World Statistics Congress - Ottawa, Canada

Predicting pollution risk using asymmetric GARCH-DCC models

Author

GP
Giuliana Passamani

Co-author

  • P
    Paola Masotti

Conference

64th ISI World Statistics Congress - Ottawa, Canada

Format: CPS Paper

Keywords: air-pollution, conditional-volatility, dynamic-correlation, time-series

Abstract

Exposure to high levels of pollution represents a major risk factor, as it may produce adverse health effects documented by numerous studies (e.g. Fuller et al. 2022). In order to work towards reaching SDG target 3.9.1, that aims to reduction in illnesses and deaths attributed to ambient air pollution, we need to define a model for measuring and predicting risk associated to exposure. Given that ambient air pollution is the outcome of complex mixtures of air pollutants emitted from various activities, an approximation of their combined effects and of impacts on health is possible if we could assume some form of independence and little correlation between the pollutants. However, there are some limitations in estimating these joint effects given nonlinear interactions among pollutants and their impacts. For the paper purpose, we consider the extension of time varying volatility models for time series data, to dynamic multivariate regression models, in which the diagonal elements of the conditional covariance matrix of the errors are modelled as univariate GARCH models, whereas the off-diagonal elements are modelled as nonlinear dynamic functions of the diagonal terms and of the conditional quasi-correlations. In other words, for measuring and predicting pollution risk we use GARCH-Dynamic Conditional Correlation (DCC) models (Engle, 2002) developed for measuring and hedging financial risk. The data set consists of daily standardized concentrations, over two years, on three pollutants, PM10, NO2 and O3, which are interrelated and represent the so-called photochemical pollution factor. The three variables are observed at a single urban monitoring site. Given the non-stationarity in the mean of the observed variables, their stochastic trends are estimated using a smooth-trend unobserved component model and we use these estimated trends to de-trend these variables to make them stationary. As observed, pollutants concentrations show the presence of significant and different GARCH effects. The objective of this paper is to explore whether the use of a multivariate asymmetric GARCH-DCC model can lead to a more accurate risk prediction for air pollution. In particular, we aim to determine how positive shocks to the observed pollutants can increase health risk. Interesting results emerge for particulate matter and ozone, both of which have great effects on human health.