Abstract:
This research highlights a multi-variety technique to examine the relationship
between dependent and independent variable in forecasting the water quality index in
Manjung Rivers and its tributaries. Multiple regression statistical method been used to
predict the water quality as the prediction is more accurate and able to be understood by
the public. The model building process been used to analyse and generate the data. There
are 4 phases under model building process and we need to run all the possible models for
each phase. For phase 1, there are 63 possible models. The number of possible model
started to reduce as we started to eliminate insignificant variable in phase 2. Then, this
model then needs to run under eight selection criteria to identify the best model for phase
3. Lastly, for phase 4 we need to identify the randomness and normality of the data. After
that, we need to verify the validity of our best model by using Mean absolute percentage
error (MAPE). According to Lewis's (1982) interpretation of MAPE results is a means to
judge the accuracy of the forecast as less than 10% is a highly accurate forecast, 11% to
20% is a good forecast, 21% to 50% is a reasonable forecast, and 51% or more is an
inaccurate forecast. Lastly, we able to identify the main parameters that give a significant
contribution to the water quality index and we able to generate a new equation in
forecasting the water quality by using the significant elements only.
