MULTIPLE LINEAR REGRESSION (MLR) MODELING OF WASTEWATER IN URBAN REGION OF SOUTHERN MALAYSIA

Abstract

A wastewater modeling contributes a valuable information in managing wastewater problems. Therefore, this study presented a multiple linear regression modeling on waste water in terms ofheavy metal concentrations (lead, copper and zinc) and water quality parameters (BOD, COD, TSS, pH, temperature). The multiple linear regression was performed eight independent variables at three selected stations in six different times. The samples were collected from waste water produced by a variety of food stalls in the urban region of Southern Malaysia. The results showed that the water pH was not influenced by lead dependent variable at all sampling stations. Pearson correlation was performed to investigate the significant relationship among continuous variables such as BOD, COD, TSS, pH, temperature, lead, copper and zinc. The 'one sample t-test' was used in order to know the significant difference between numerical variables in all stations during the sampling period. Multiple comparisons and Fisher's LSD method were also implemented for investigating the differentiation ofinterested numerical variables.