Successive days of precipitation are known to cause flood in
monsoon-type countries. Forecasting of daily precipitation helps
us to predict the occurrences of rainfall and number of wet days.
By using the maximum five-day accumulated precipitation, we can
predict the magnitude of precipitation in a specific period as
it may indicate the extreme precipitation. In this study, a method
to forecast monthly extreme precipitation using artificial neural
networks (ANN) was assessed by using past data of the maximum
accumulated five-day precipitation (MX5d) and global climate
indices such as Southern Oscillation Index (SOI), Madden Jullian
Oscillation (MJO) and Dipole Mode Index (DMI) in Kuantan and
Kota Bharu, Malaysia. The use of combined inputs (MX5d with SOI,
MJO, and DMI) is proposed to investigate the concurrent effect
of lagged values of local precipitation data and global climate
indices on seasonal extreme precipitation. Four cases of data
were sampled which represent two major seasonal variations in
Malaysia. The analysis of extreme precipitation trend is important
for future prediction of high precipitation events in the area
of interest. ANN is widely applied in the hydrology field due
to non-linear ability in predicting non-stationary and seasonal
data. Here, we have compared ANN with seasonal autoregressive
integrated moving average (ARIMA)) and regression analysis using
out-of-sample test data. The results for Kuantan showed that
seasonal ARIMA is the best method to forecast extreme precipitation
when MX5d lags were used as inputs. For Kota Bharu, ANN has better
generalisation ability than ARIMA and regression analysis when
dual inputs (lagged MX5d and lagged global climate indices) were
utilized in the model.