Coupling of quantile regression into boosted regression trees (BRT) technique in forecasting emission model of PM10 concentration

Abstract

Air pollution is currently becoming a significant global environmental issue. The sources of air pollution in Malaysia are mobile or stationary. Motor vehicles are one of the mobile sources. Stationary sources originated from emissions caused by urban development, quarrying and power plants and petrochemical. The most noticeable contaminant in the Peninsular of Malaysia is the particulate matter (PM10), the highest contributor of Air Pollution Index (API) compared to other pollution parameters. The aim of this study is to determine the best loss function between quantile regression (QR) and ordinary least squares (OLS) using boosted regression tree (BRT) for the prediction of PM10 concentration in Alor Setar, Klang and Kota Bharu, Malaysia. Model comparison statistics using coefficient of determination (R2), prediction accuracy (PA), index of agreement (IA), normalized absolute error (NAE) and root mean square error (RMSE) show that QR is slightly better than OLS with the performance of R2 (0.60–0.73), PA (0.78–0.85), IA (0.86–0.92), NAE (0.15–0.17) and RMSE (9.52–22.15) for next-day predictions in BRT model.

Does Rural Production–Living–Ecological Spaces Have a Preference for Regional Endowments? A Case of Beijing-Tianjin-Hebei, China

“…Contributions of independent variables and the marginal effect within the value range are clearly shown in the model results [59,60]. So far, BRT model has been successfully applied in urban heat island factor research [61,62], ecological risk assessment research [57,63] and many other fields, with high value.…”

Prediction of Daily Air Pollutants Concentration and Air Pollutant Index Using Machine Learning Approach

“…The highest value of R 2 was ranked as 1, the second highest value was ranked as 2, and the lowest was ranked as 3. Then, the ranked values for RMSE, NAE, and R 2 were summed up to find the lowest total ranking values among the three models, which indicate the best (Shaziayani et al, 2021).…”

“…Furthermore, there are many studies on the prediction of air pollutants concentration. However, most studies only focus on predicting one pollutant or a few pollutants only such as (Hamid et al, 2017;Shaadan et al, 2019;Alias et al, 2021;Shaziayani et al, 2021), and there is limited study on predicting all pollutants concentration in a study. Thus, this study will predict the concentration of all major air pollutants: ozone, carbon monoxide, nitrogen dioxide, sulfur dioxide, and particulate matter less than 10 micrometers (PM 10 ).…”

Predicting Particulate Matter (PM10) during High Particulate Event (HPE) using Quantile Regression in Klang Valley, Malaysia

“…It is progressively developing as a thorough method to the statistical analysis of linear and nonlinear models [11]. The QR represented the non-central location of a distribution which allow the approach to be more useful and precise [12]. [13] concluded that QR models inhibit some advantages compared to MLR since it does not rely on any properties, is independent or only mildly dependent, is robust to outliers, and is distribution free.…”