Forecasting of PM10 Using Autoregressive Models and Exponential Smoothing Technique

Abstract

Particulate matter with 10 μm or less in diameter (PM 10 ) have adverse effects on environment and human health. To reduce PM 10 emissions in India, it is essential to have models that accurately estimate and predict PM 10 concentrations for reporting and monitoring purposes. In this paper Exponential Smoothing Technique and Autoregressive (AR) models are developed to forecast 1-month ahead value of PM 10 for Allahabad city which is novelty of this study. AR (1) and AR (5) models are developed using Burge and Yule Walker methods. The mean absolute percentage error (MAPE) for Burge method in AR (1) and AR (5) are 14.23% and 10.20%. The MAPE for Yule Walker method in AR (1) and AR (5) are 32.72% and 31.13%. The MAPE in Exponential Smoothing Technique is 5.81% which shows it forecasts better than AR model based on Burge and Yule Walker methods. It is found that Burge Method in AR (5) has less MAPE than Yule Walker Method. Therefore Exponential Smoothing Technique can be used to forecast PM 10 for cities in India, showing it is beneficial for giving prior information for human health.

Prediction of optimum CNC cutting conditions using artificial neural network models for the best wood surface quality, low energy consumption, and time savings

“…The MAPE value, which is frequently used by researchers to evaluate ANN model performances, is expected to be below 10% (Antanasijević et al 2013;Tiryaki et al 2016). It has been demonstrated that the prediction performance of ANN models is high with these values lower than 10% (Yadav and Nath 2017). It is stated in the literature that it is extremely important to calculate RMSE values as well as MAPE values in order to determine the performance of prediction models (Kucukonder et al 2016).…”

The use of an artificial neural network for predicting the gloss of thermally densified wood veneers

“…The MAPE values were 2.856% for the training set and 8.556% for the testing set. Several researchers reported that model performance is accepted as excellent if MAPE ≤ 10% (Chang et al 2007, Aydin et al 2014, Yadav and Nath 2017. Accordingly, the prediction ability of the ANN model can be accepted as excellent since its MAPE value is lower from 10%.…”

Estimation of Formaldehyde Emissions of Plywood Treated With Fire Retardant Chemicals by Artificial Neural Network

“…MAPE, en önemli değerlendirme kriterlerinden biridir ve birçok araştırmacı, MAPE'yi kullanarak model performansını değerlendirmiştir (Antanasijević et al, 2013;Tiryaki et al, 2016, Yadav & Nath, 2017. Literatürde MAPE değerinin%10'un altında olması durumunda model performansının yüksek olduğu belirtilmiştir (Yadav & Nath, 2017).…”

“…MAPE, en önemli değerlendirme kriterlerinden biridir ve birçok araştırmacı, MAPE'yi kullanarak model performansını değerlendirmiştir (Antanasijević et al, 2013;Tiryaki et al, 2016, Yadav & Nath, 2017. Literatürde MAPE değerinin%10'un altında olması durumunda model performansının yüksek olduğu belirtilmiştir (Yadav & Nath, 2017). Bu çalışmada, retensiyon miktarına ait MAPE değerleri, eğitim verileri için %1.630 ve test verileri için %4.372 olarak hesaplanırken, formaldehit emisyon değerlerine ait MAPE değerleri, sırasıyla %0.891 ve %3.529 olarak belirlenmiştir (Tablo 1 ve 2).…”