Modelling Indoor Air Quality by Selecting Input Parameters Amalina Abu Mansor (1), Samsuri Abdullah (2*), Aimi Nursyahirah Ahmad(2), Marzuki Ismail(1)
(1)Faculty of Science and Marine Environment, Universiti Malaysia Terengganu, 21030, Kuala Nerus, Terengganu, Malaysia
(2)Faculty of Ocean Engineering Technology and Informatics, Universiti Malaysia Terengganu, 21030, Kuala Nerus, Terengganu, Malaysia
Abstract
We assess the indoor air quality parameters by monitoring chemical and physical variables at primary school in Terengganu. The data were collected during school hours for modeling purposes. Three models were developed including the stepwise method (Model A), correlation analysis (Model B) and enter method (Model C). It is found that Model A outperformed Model B and C due to high Coefficient of determination (R2) of Model A (R2=0.703) compared to Model B (R2=0.697) and Model C (R2=0.187). The ability of the method to manage large amounts of potential predictor variables, fine-tuning the model to choose the best predictor variables from the available options is the main reason for the best selected method.
Keywords: temperature, indoor air quality, regression, school
