able 4 shows the fitting results
and effect on climate variables including
parameter and significant value of Model
1 and Model 2. Significance on interaction
effect between maximum or minimum
monthly temperature, RH and wind speed
with the number of malaria incidences
are concluded.
T he results show that Model 2 has
the best accuracy, therefore the prediction
process model 2 is used. Both models are
adapted by non linear mixed-regression
technique; Model 2 used interaction of
climate variables, Model 1 only use
climate variables at max lag period (τmax).
As a result, Model 2 which combined the
interaction between climate variables
shows potential better prediction than
Model 1. In fact, the regression model in
this study can represent and predict malaria
burden with reasonable accuracy when
there is sufficient malaria epidemiological
data (non under reported province data)
available for input to the model. In the
prediction period, although there may be
some over estimated between the actual
incidences and predicted, The RMSE
(Table 5) is relatively small when compared
with overall malaria burden (Figure 1).