Ari Pani Desvina, - and Arinal Haque, - and Riswan Efendi and Muspika Hendri, - and Mas’ud Zein, - and Sri Murhayati, - (2018) Air Pollution Prediction with Hotspot Variable based on Vector Autoregressive Model in Pekanbaru Region. Proceedings of the International Conference on Mathematics and Islam (ICMIs 2018) . SCITEPRESS – Science and Technology Publications, Lda. ISBN 978-989-758-407-7

Preview

Text Air_Pollution_with_Hotspot_Variable.pdf Download (869kB) | Preview

Abstract

The air quality is widely caused by pollution of particulate matter (PM10) and meteorological elements. For examples, rainfall, solar radiation, air temperature, humidity, wind velocity, and hotspot. In analysis data (ADV), the used variables are more than one variable, so that the best model for modeling and forecasting multivariate data is vector autoregressive (VAR). The VAR model is chosen because it is one of multivariat analysis for time series data and it is able to describe the interconnectedness among variables. The aim of this research is to find the best model for PM10 concentrations with other meteorological elements in Pekanbaru by using VAR model, and to determine the prediction result of PM10 concentration in the future. Furthermore, the monthly data of Pekanbaru region from January 2011 until December 2015 was used for training and testing. The result showed the best model for predicting PM10 is VAR(1). It can be summarized that rainfall, solar radiation, humidity and hotspot variables have been interconnected with PM10. Based on proposed model, the concentration of PM10 data increased from January 2016 until December 2017

Actions (login required)

View Item