Estimating the Mean of PM2.5 with Missing Data in the Area Around Electricity Generating Authority of Thailand Using the Improved Compromised Imputation Method

Abstract

Particulate matter with an aerodynamic diameter of less than 2.5 m or , is one of the air pollutants that has been found to be at unsafe levels for a number of years in Thailand, leading to public health concerns. In order to lessen the detrimental effects of air pollution, monitoring and analysis of concentration are crucial. Following the study of data from the Pollution Control Department Report in the area around the Electricity Generating Authority of Thailand in January 2019, it was found that there was data missing in the information. It is well-known that missing data can reduce the accuracy of data analysis. To solve the missing data problem, this paper proposes an improved method of compromised imputation and a corresponding resultant estimator to deal with estimating the mean of concentrations in the area. The bias and mean square error of the estimator obtained from the proposed method were derived. The conditions which favor the performance of our estimator over other estimators obtained from the mean, ratio, and compromised imputation methods were obtained using mean square error to apply in the area. The mean of concentrations in this case using the proposed estimator was equal to 47.13 g/m3, indicating that it did not exceed unsafe levels (<50 g/m3) under certain conditions. In order to support more accurate data analysis that will lead to effective management of air pollution problems in the future, this research proposes a new method that is more effective than the existing methods under missing data problem.