Page 429 - Contributed Paper Session (CPS) - Volume 6
P. 429
CPS2008 Syafrina A.H et al.
models are used namely Gamma and Weibull distributions. The shape and
scale parameters for both distributions were estimated by using MLE. Different
criteria of GOF had been used to test whether the model suits for the data by
referring to the hypothesis testing. CvM and K-S tests as the methods of GOF
with the hypothesized distributions, Gamma and Weibull distributions.
Based on the result of K-S test, both distributions are not suitable for the
data with both p-values are less than 0. For another test, which is CvM test,
the result shows that Gamma and Weibull distributions are fitted to the rainfall
data with the p-values 0.3288 and 0.3206, higher than 5% significance level. In
summarise, Weibull distribution is the best model to be fitted to the daily
rainfall data at Penang International Airport based on the minimum of GOF
tests (the highest statistical test score). For further study, must be focused on
the performance comparison of modified distributions such as Mixed Gamma
and Mixed Weibull distributions since daily rainfall amount for the majority
states in Malaysia is very well represented using two components. Moreover,
monsoon period also can be considered in the future study since rainfall
amount in Malaysia is affected by monsoons season. This can be done by
separating the time period by month according to the monsoon period. This
indicated that the rainfall distribution in Malaysia is consists of minimum and
maximum rainfalls. Other potential work includes of using variety of GOF test
such as Anderson Darling and Chi Square tests. This may help to predict the
most suitable model much precise based on the majority of the minimum GOF
tests. Moreover, more stations from each state in Malaysia should be used to
perform this study in future since different station of rainfall amount in
Malaysia is affected by different monsoon season. This is due to the
geographical of the location.
References
1. Abdullah, M. A., & Al-Mazroui, M. A. (1998). Climatological study of the
southwestern region of Saudi Arabia. I. Rainfall analysis. Climate
Research, 9(3), 213-223.
2. Deidda, R., & Puliga, M. (2006). Sensitivity of goodness-of-fit statistics to
rainfall data rounding off. Physics and Chemistry of the Earth, Parts
A/B/C, 31(18), 1240-1251.
3. Duan, J., Selker, J., & Grant, G. E. (1998). Evaluation of probability density
functions in precipitation models for the Pacific Northwest. Journal of the
American Water Resources Association, 34(3), 617627.
4. Husak, G. J., Michaelsen, J., & Funk, C. (2007). Use of the gamma
distribution to represent monthly rainfall in Africa for drought monitoring
applications. International Journal of Climatology: A Journal of the Royal
Meteorological Society, 27(7), 935-944.
418 | I S I W S C 2 0 1 9