Search

It is noting that the response variable and the explanatory variables are highly correlated in high dimension data. Hence, the selection of informative variables is important in order to achieve a better model interpretation and concomitantly improve the accuracy of the prediction. In this study, the variable selection in stock market price using statistical approach was carried out. It is pertinent since most of the previous study only concerns on the financial interests of the stock market. Therefore, this study considers the homogeneity structure in the highly correlated data on yearly stock market price by applying ordered homogenous pursuit lasso (OHPL) method. The performance results of OHPL were compared with lasso and elastic net. As a result, OHPL a had higher number of selected variables and a better prediction power than of lasso and elastic net. In conclusion, OHPL shows its capability to enhance variable selection while increasing the prediction power of the selected variables than its counterpart. [ABSTRACT FROM AUTHOR]

Portfolio investment is one of the important topics in financial management. Its aim of having the wealth in a period of time with maximum return and minimum risk has been well-defined in the literature. Specifically, the portfolio selection is arisen to be an interesting investment issue. Since the risk is defined as the probability of losses relative to the expected return in the portfolio investment, the existing risk would influence the rate of return and affect the investment decision. In this paper, the portfolio selection problem is discussed. The aim is to provide an effective quantitative study in managing the portfolio. In the beginning, the rates of return are calculated and it is followed by computing the expected rates of return and the related covariance matrix. Accordingly, the mean-variance optimization model, which is the quadratic programming model, is formulated and solved by the quadratic programming technique. Here, the objective function to be minimized represents the risk of the portfolio selection, while the weights of the portfolio selection, which are defined as the decision variables, are normalized with the total amount of investment, and the expected return is assigned to a certain targeted return. For illustration, a set of historical data on the stock prices in the Financial Times Stock Exchange (FTSE) Bursa Malaysia Kuala Lumpur Composite Index (KLCI) is considered. Through simulation and optimization of the mean-variance portfolio model, the result shows the relationship between risk and return with the various targeted returns. Moreover, by applying the Sharpe ratio, the optimal weight of the portfolio selection, which is the optimal portfolio decision, is properly suggested. In conclusion, the simulation optimization model for the portfolio selection problem is highly recommended. [ABSTRACT FROM AUTHOR]

The field of survival analysis has made a tremendous progress during the latter half of the twentieth century. Weibull distribution and parametric models (proportional hazards models) have played important roles in the analysis of survival data. In this paper, a two-parameter modified Weibull distribution, three-parameter modified Weibull distri- butions and a four-parameter modified Weibull distribution are demonstrated that can be adopted in the proportional hazards models. Smoothed hazard estimate is used to indicate that Weibull and modified Weibull models are preferred to apply in the coronary heart disease (CHD) data. Weibull and modified Weibull proportional hazards models are fitted to the CHD data. It is found that the two-parameter modified Weibull proportional hazards model has a better fit to the CHD data than Weibull and other modified Weibull proportional hazards models. [ABSTRACT FROM AUTHOR]

Everybody has at least a smartphone and the technology keep on evolving from time to time, same goes to mobile applications. Mobile applications are currently more towards learning, business and entertainment purposes, but why not we extend the use of the technology to forensic investigation? In this paper, we are highlighting the potential of a mobile application on smartphones to support investigation in crime scenes. As we know, crimes involving firearms has been a threat to humanity since time immemorial. The mass production of handguns and other firearms since the nineteenth century has similarly expanded crime numbers globally. Thus, it is critical to identify the firearm used whenever a firearm crime has been reported, which calls for expertise in forensic ballistics. Firearms identity is a crucial goal of firearm analysis. From previous research, the firing pin impact impression on a cartridge case from a fired bullet is one of the most extensive clues in firearms identification. Since nowadays a lot of crime cases involve guns, a portable device should help the police and the forensic teams for immediate firearm identification, whereby they can pre-analyze the firearm impression on the spot to gain some prior information, which is cost, space and time saving. In this paper, mobile application for firearm identification is proposed, named Firearm Identification Mobile Application (FIMA). This firearm identification application will analyze numerical features extracted from the firing pin impression image as the unique features, and immediately catalogues the features using the backpropagation neural network classification technique. It is found that 'trainlm' performed the best compared to the other training algorithms based on average overall correct classification rates for all features. On the other hand, 'trainscg' performed the best based on centre firing pin impression images. In the near future, a portable firearm analysis device will be developed, namely Portable Firearm Analysis Device (PAFAD) to further assist in forensic investigation activities. [ABSTRACT FROM AUTHOR]

Income inequality is a matter of concern for any society and an interesting discussion topic, particularly for developing countries, since high-income inequality could be detrimental to economic growth. In Malaysia, upper-income earners are classified as the T20 group. This study analyzes the demographic profile of members of the T20 group living in urban areas in Malaysia. The upper-income distribution is described using the Pareto model, and a robust and efficient estimator called the probability integral transform statistic estimator is utilized to estimate the shape parameter of this model. Thereafter, a Gini index and Lorenz curve are derived to analyze the severity of income inequality among T20 group members. [ABSTRACT FROM AUTHOR]

The association between dietary habits and cardiovascular diseases (CVDs) has been established. However, the dietary habits vary among countries, therefore, there is scarcity in the literature for a standardized instrument to assess the dietary habits. This paper is aimed to evaluate the psychometric properties of modified dietary habits instrument, originally based on a food frequency questionnaire (FFQ). In this case-control study, it was used to find the most prevalent dietary habits in CVDs patients and their corresponding controls. The modified instrument consisted of 13 items representing thirteen food groups identified from CVDs related studies and FFQ. A sample of 180 (90 cases and 90 controls) was selected from the largest cardiac hospital of Pakistan through systematic sampling. Data was collected through modified instrument using direct investigation method. Twelve food groups were validated which further constituted two main factors labelled; healthy and unhealthy dietary habits. Exploratory factor analysis explained 70% and 68% variation through these two main identified dietary habits in cases and control, respectively. Internal consistency reliability for healthy and unhealthy dietary habits in cases were 0.905 and 0.879 and almost similar coefficient of reliability was recorded in the control group. Further, the mean score of healthy dietary habits was significantly less than unhealthy dietary habits (p<0.001) in the cases and vice versa in controls. This study concludes that the modified instrument is valid and reliable and should be used to assess the dietary habits in CVDs related studies. Further, mean score of unhealthy dietary habits can help in disease prediction models and early diagnosis of cardiac problems. [ABSTRACT FROM AUTHOR]

A copula-based analysis is usually performed in hydrology for the purpose of water management and flood prevention by analyzing the hydrological attributes, such as rainfall volume and duration. The primary goal of this study is to fit the rainfall volume and duration into copulas. The analysis was conducted on six selected rainfall stations in Malaysia with hourly rainfall data. The correlation between volume and duration is measured by Kendalls correlation coefficient. Clayton, Gumbel-Hougard (GH), Frank and Joe copulas of Archimedean family are used and the parameter of the copulas is estimated using Maximum Likelihood Estimation (MLE). Akaikes Information Criterion (AIC), three graphical methods, and Goodness-of-Fit (GOF) tests are used to find the best parameter estimation of copulas. The three graphical methods used are the conditional distribution function of Y|X, the distribution function of the copula, and a nonparametric estimate of the distribution of the copula approach. The GOF tests are The Kolmogorov-Smirnov (K-S) and Chi-Square (χ2) tests. Results estimate that Frank copula is found to be suitable to represent the rainfall volume and duration for all stations. [ABSTRACT FROM AUTHOR]

Seabed Logging (SBL) is a technique that employs high-powered electric dipole source to emit electromagnetic (EM) signal to detect hydrocarbon (HC) reservoirs beneath the seabed. This application is based on electrical resistivity contrasts between target reservoirs and its surrounding. SBL analysis can become a challenging task when the target reservoirs are thin, and the contrasts of resistivity are not very significant. As HC reservoirs are getting thinner, target responses and reference (non-HC) responses are difficult to be distinguished. Addressing this problem, we propose a simple statistical method, Gaussian Process (GP), to model one-dimensional (1-D) SBL data with uncertainties quantification. In this paper, Computer Simulation Technology (CST) software was used to replicate SBL models with five different thicknesses of HC. Some characteristics of the SBL models such as seawater depth, reservoir thickness and reservoir depth were imitated as the case study of Troll West Oil Province, North Sea. We developed 1-D forward GP model for all the SBL responses. Both modelled responses, target and reference, were compared and mean percentage differences between the responses were then calculated. For every comparison, confidence bars for each modelled response were observed to confirm the existence of thin HC. For model validation, root mean square errors (RMSEs) between modelled and generated (CST software) data were calculated. The confidence intervals revealed that the target and reference responses are distinguishable for all HC thicknesses, and the calculated RMSEs showed that GP is reliable to be applied in SBL data to provide uncertainties quantification. [ABSTRACT FROM AUTHOR]

Big Data phenomenon has risen questions concerning the capabilities of classical methods of statistical inference, most of which were rendered obsolete. This problem exists in the fields of geostatistics for a long time but have not been properly addressed. Adaptive Spatial Compression Clustering (ASDC), a new spatial data compression method, has done well where methods using full dataset prove futile. Presence of global outliers in the spatial data affects the formation of spatial dispersion function which subsequently affects the outcome of the spectral clustering; this, in effect, affects spatial contiguity. With the aim of neutralizing the effect of the global outliers, absolute difference of all pairs of spatial observations is used. Hampel's outlier test was adopted before taking local spatial dispersion. Simulation results of synthetic spatial fields reveal worthwhile in treating the effect of outliers. [ABSTRACT FROM AUTHOR]

There were many cases reported for ovarian cysts and it was difficult to recognize the size of cyst whether it is a large or a small cyst. Thus, this research was conducted to predict the size of benign cysts according to the patient's symptoms and its risk factors, to analyze the relationship between the size of benign cyst and the type of surgical procedures that the patient needs to undergo and to obtain which variable gives most important effect to the model. Retrospective observation study has been applied in this research since the data was collected from the monthly census file for year 2016 in Ward of Obstetrics and Gynecology from Hospital A, Johor Bahru. From the 89 patients of benign ovarian cysts, only 26 of the patient have small cysts and the other 63 patients have large benign cysts. From analysis, it found that the significant symptoms towards the predicting of Size of Benign Cyst were Abdominal Pain and Intermenstrual Bleeding. There were no significant for the risk factors. There was a significant relationship between the Size of Benign Cysts and the Type of Surgical Procedure. From the association, it can be concluded, if the patient had a large size of cysts, they will need to undergo laparotomy surgery. By using log-linear analysis, it is found that the most important effect in model was Type of Operation. [ABSTRACT FROM AUTHOR]

Undoubtedly, early detection and characterization of brain tumor is critical in clinical practices. Automated diagnosis using neuroimaging tool like MRI guided by machine learning approaches has been the focus of numerous researches. In this study, various feature extraction, dimensionality reduction and supervised classification models are explored, evaluated and compared under different finite number of features to identify the optimal pathway/pipeline for classification of types of brain tumor, namely meningioma, glioma and pituitary tumor. The performance metrics utilized include accuracy, Kappa statistic, sensitivity, precision, F-measure, training time and test time. Results show that RBF SVM (pairwise coupling) under 80 PLS features achieved the highest average accuracy (95.02% ± 0.19%) among all other machine learning pipelines. [ABSTRACT FROM AUTHOR]

Unintentional poisoning is on the rise globally and thus a statistical analysis has been conducted. This study focuses on implementing a binary logistic regression analysis to study the factors that lead significantly to unintentional and intentional poisoning in Malaysia. The data consists of information from 592 patients with 7 predictor variables and a binary response variable. Preliminary analysis and chi-square test are conducted before performing the analysis. To obtain a good logistic model, there are three main steps: model building, diagnostic checking and drawing inferences from the parameters. The model provides parameter inferences such as odds ratio and estimation of predicted probabilities. Based on the results, unintentional poisoning is significantly influenced by physical form of poison, type of natural toxin and exposure route. Odds ratio estimation indicates that patients exposed to poisoning through bites or stings are more likely to have unintentional poisoning compared to other exposure routes. Generally, estimation of predicted probability are high for unintentional poisoned patients when they are bitten by animals. [ABSTRACT FROM AUTHOR]

Workforce size problem is a common issue in manufacturing, which is to determine the optimal number of workers required such that the productivity of a product is ensured. As such, the situation of understaffing and overstaffing in the manufacturing could be handled properly, in turn, to save the labour cost and to fulfil the economic supply and demand. In this paper, the workforce size problem in manufacturing is described, and the dynamic programming approach is applied to solve the workforce size problem. In the discussion, the return function of the workforce size problem and the state transition equation are identified. Then, the backward recursion equation of dynamic programming is formulated. According to each stage considered, the possible states are determined, and the related value of the return functions are calculated. It is followed by establishing the optimal decision policy for having the optimal number of workers. For illustration, an example of the workforce size problem in manufacturing, which is on a weekly basis, is studied. The result obtained showed that the minimum recruitment cost is RM33, where the optimal number of workers for the respective weeks is four workers, eight workers, eight workers, six workers, and six workers. In conclusion, the efficiency of the dynamic programming method in solving the workforce size problem in manufacturing is highly demonstrated. [ABSTRACT FROM AUTHOR]

The parametric method is often applied to model income data and to assess income inequality. Parametric estimation of income distribution is convenient because it facilitates subsequent inferences about the measured inequality. In this study, we conduct a comparative assessment of four parametric distribution candidates in terms of model fitting for the income of poor households in Malaysia on the basis of survey data of household incomes for the years of 2007, 2009, 2012 and 2014. The considered models involve the reverse Pareto, shifted reverse exponential, shifted reverse stretched exponential and shifted reverse lognormal. We also derive the explicit expression of Lorenz curve and Gini coefficient for all four considered models. Finally, on the basis of the best parametric model, income inequality analysis of poor households in Malaysia is assessed using Lorenz curve and Gini coefficient. [ABSTRACT FROM AUTHOR]

This study aims to analyse the trend of Malay voters' support for Barisan Nasional during the Malaysian General Elections which spans from year 1995 until year 2018. Data of registered Malay voters during the respective general election at the Parliament level is obtained via a secondary source, which is the Malaysia Election Commission's portal. The pattern of support from the Malay voters for Barisan Nasional shows a diminishing trend over the years which shows that there is decline of support from Malay registered voters for Barisan Nasional over the election years. From the regression analysis, the results show that there is a significant positive relationship between the percentage of registered Malay voters and the percentage of votes for Barisan Nasional's party. This results suggest that the support of Malay voters is crucial for Barisan Nasional to win the election at the Parliamentary level. [ABSTRACT FROM AUTHOR]

This paper presents the application of simulation in modeling the food manufacturing system at XYZ factory. The problems faced by the factory are the bottlenecks that happen in their food manufacturing system which consist of four processes i.e. rolling and fermentation process, drying process, sorting and grading process and packaging process. From the data collection provided by the management of the company showed that the drying process and packaging process has high waiting processing times and lead to the entities from the previous process waiting for a long time. Moreover, the imbalance workers for each process can affect the productivity and output of the production. Simulation method was used for modeling the system. The results of the simulation model are analyzed for the suggestion of improvement. The scoring model was used in this study to select the best improvement model from nine improvement models that have been developed. For the scoring model, the company has selected five criteria. The criteria are total production, average worker utilization, and average total production time, the average number of entities in the system and total number of workers involved in the manufacturing system. Each of the criteria has been weighted by the company in terms of its relative importance. In this study, IP4 is the best improvement model where the model achieves the highest score in the scoring model and can maximize the total production and average resource utilization. IP4 has recommended the reallocation of workers. The application of the simulation method and the results from this research can help the management of the factory to make better decisions thus enhance the food manufacturing system. [ABSTRACT FROM AUTHOR]

In this note, we consider a supersingular integral equations (SuperSIEs) of the first kind on the interval [−1,1] with the assumption that kernel of the hypersingular integral is constant on the diagonal of the domain D = [−1,1]×[−1,1]. Projection method together with Chebyshev polynomials of the first, second, third and fourth kinds are used to find bounded, unbounded and semi-bounded solutions of SuperSIEs respectively. Exact calculations of singular integrals for Chebyshev polynomials allow us to obtain high accurate approximate solution. Gauss- Chebyshev quadrature formulas are used for high accurate computations of regular kernel integrals. Two examples are provided to verify the validity and accuracy of the proposed method. Comparisons with other methods are also given. Numerical examples reveal that approximate solutions are exact if solution of SuperSIEs is of the polynomial forms with corresponding weights. It is worth to note that proposed method works well for large value of node points and errors are drastically decreases. [ABSTRACT FROM AUTHOR]

The study on probability theory in finite rings has been an interest of various researchers. One of the probabilities that has caught their attention is the probability that two elements of a ring have product zero. In this study, the probability is determined for a finite ring R of matrices over integers modulo four. First, the annihilators of R are determined with the assistance of Groups, Algorithms and Programming (GAP) software and then the probability is calculated using the definition. Next, by using the results obtained, the zero divisor graph of the ring R is constructed. A zero divisor graph is defined as a graph in which the zero divisors of R are its vertices and two vertices are connected by an edge if their product is zero. [ABSTRACT FROM AUTHOR]

In biology, DNA splicing system models the recombination behaviours of DNA molecules which require the presence of restriction enzymes. Since it is quite incompetent to describe DNA splicing system in one-dimensional string, hence graph theory is used in describing the complexity of DNA splicing system. Thus by graph splicing system, a type of semigraph can be obtained, called as n-cut spliced semigraph. Various concepts of graph theory have been applied on semigraph to study its characteristics and properties. In this study, the focus is on the application of the folding technique on n-cut spliced semigraph.It is observed that the maximum number of folding for every n-cut spliced semigraph is n+1 times. Also, the final form of new semigraphs obtained are determined. [ABSTRACT FROM AUTHOR]