This paper proposes A Hybrid Wavelet-Auto-Regressive Integrated Moving Average (W-ARIMA) model to explore the ability of the hybrid model over an ARIMA model. It combines two methods, a Discrete Wavelet Transform (DWT) and ARIMA model using the Standardized Precipitation Index (SPI) drought data for forecasting drought modeling development. SPI data from January 1954 to December 2008 used was divided into two - (80%/20% for training/testing respectively). The results were compared with the conventional ARIMA model with Mean Square Error (MSE) and Mean Average Error (MAE) as an error measure. The results of the proposed method achieved the best forecasting performance.
This journal renders the random walk behaviour of the Malaysian daily share return, through tests of efficient market hypothesis (EMH) based on three different financial periods, namely growth, financial crisis, and recovery period. This review also covers the behaviour of extreme return for weekly and monthly series generated from Block maxima-minima method. Autocorrelation Function test (ACF) and Ljung-Box test had been employed to measure average correlation between observations, while Augmented Dickey-Fuller (ADF), Phillips-Perron (PP), Kwiatkowski Phillips Schmidt Shin (KPSS) test had been used to scan the unit root and the stationarity. Multiple variance ratio tests had also been conducted to examine the random walk behaviour. Serial correlation test indicated that the movement of daily return during the financial crisis period was weak-form efficiency. The unit root and stationary tests suggested that each daily series was stationary, but trend stationary for extreme cases. Variance ratio tests indicated that the return during the recovery period was weak-form inefficiency due to the short lag autocorrelation in series.
Identification of the true statistical distributions for various hydrologic data sets is a major problem facing engineers. The four-parameter kappa distribution is a combination of the established distribution including the Generalised Extreme Value (GEV), Generalised Logistic (GL), Generalised Pareto (GP) and the Gumbel distribution were considered in this study. The main objective of this study was to develop the method of LQ-moments for the kappa distribution. The performance of the LQ-moments was compared with L-moments through eight problems using published data sets. The results showed that the performance of both methods, the LQ-moments and L-moments worked equally well.
Sejak taburan normal ditemui dan ianya merupakan salah satu taburan yang penting dalam statistik, terdapat banyak pengujian statistik yang dibangunkan untuk menguji kenormalan data. Namun begitu masih tidak banyak kajian yang dilakukan untuk melihat kembali keupayaan pengujian statistik yang sedia ada. Sebahagian daripada pengujian statistik didapati mudah tetapi hanya sesuai untuk sesuatu keadaan. Dalam kajian ini, pengujian statistik berdasarkan statistik Cramer-von Mises cuba diperbaiki berdasarkan rumus Weibull. Kekuatan statistik yang baru ini dibandingkan kekuatan dengan statistik traditional Anderson-Darling (AD), Cramer von-Mises (CR), Kolmogorov-Smirnov (KS) dan Shapiro-Wilk (SW). Kajian simulasi berdasarkan beberapa taburan yang berbeza menunjukkan pengujian statistik yang dicadangkan paling sesuai untuk menguji kenormalan.
Kaedah momen merupakan salah satu teknik termudah dan sering digunakan dalam statistik hidrologi. Bagaimanapun, penganggar momen selalunya memberikan kualiti yang rendah dan tidak sebaik penganggar L-momen terutamanya untuk taburan dengan tiga parameter atau lebih. Pada masa kini, banyak kajian analisis frekuensi serantau dalam hidrologi menggunakan gambar rajah nisbah L-momen untuk memilih taburan yang sesuai bagi data hidrologi dan meteorologi. Kelebihan utama gambar rajah nisbah L-momen adalah pemilihan beberapa taburan yang sesuai boleh dilakukan menggunakan satu graf sahaja. Tujuan utama kertas ini untuk melihat kembali LQ-momen dan untuk membangunkan gambar rajah LQ-momen berdasarkan penganggar median. Menggunakan data aliran banjir dari 73 buah stesen dalam Semenanjung Malaysia, kami meninjau kesesuaian pelbagai model kebarangkalian menggunakan gambar rajah nisbah LQ-momen dan dibandingkan dengan gambar rajah nisbah L-momen. Hasil kajian menunjukkan gambar rajah nisbah LQ-momen secara umumnya memberikan keupayaan yang baik sebagaimana gambar rajah nisbah L-momen dalam memilih taburan frekuensi dan ini membuatkan ianya sesuai untuk dijadikan sebagai pilihan yang menarik untuk digunakan dalam analisis frekuensi banjir.
Knowledge related to distributions of rainfall amounts are of great importance for the design of water related structures. The greater problem facing hydrologists and engineers identifying the best distribution form for regional data. The main goal of the study is to perform regional frequency analysis of maximum daily rainfalls selected each year among daily rainfalls measured over stations in Selangor and Kuala Lumpur by using the L-moment method. Several distributions were taken into account in this study which include two-parameter normal (NOM), lognormal (LN2), three-parameter lognormal (LN3), logistic (LOG), generalized logistic (GLO), extreme value type I (EV1), generalized extreme value (GEV) and generalized Pareto (GPA) distribution. The most suitable distribution was determined according to the mean absolute deviation index (MADI), mean square deviation index (MSDI) and the L-moment ratio diagram. The result of this study showed that the GLO distribution is the most suitable distribution to fit the data of maximum daily rainfalls for stations in Selangor and Kuala Lumpur.
Since rice is a staple food in Malaysia, its price fluctuations pose risks to the producers, suppliers and consumers. Hence, an accurate prediction of paddy price is essential to aid the planning and decision-making in related organizations. The artificial neural network (ANN) has been widely used as a promising method for time series forecasting. In this paper, the effectiveness of integrating empirical mode decomposition (EMD) into an ANN model to forecast paddy price is investigated. The hybrid method is applied on a series of monthly paddy prices from February 1999 up to May 2018 as recorded in the Malaysian Ringgit (MYR) per metric tons. The performance of the simple ANN model and the EMD-ANN model was measured and compared based on their root mean squared Error (RMSE), mean absolute error (MAE) and mean percentage error (MPE). This study finds that the integration of EMD into the neural network model improves the forecasting capabilities. The use of EMD in the ANN model made the forecast errors reduced significantly, and the RMSE was reduced by 0.012, MAE by 0.0002 and MPE by 0.0448.