Aderemi Oluyinka Adewumi

Comparison of ARIMA and Artificial Neural Networks Models for Stock Price Prediction

This paper examines the forecasting performance of ARIMA and artificial neural networks model with published stock data obtained from New York Stock Exchange. The empirical results obtained reveal the superiority of neural networks model over ARIMA model. The findings further resolve and clarify contradictory opinions reported in literature over the superiority of neural networks and ARIMA model and vice versa.

On the Performance of Linear Decreasing Inertia Weight Particle Swarm Optimization for Global Optimization

Linear decreasing inertia weight (LDIW) strategy was introduced to improve on the performance of the original particle swarm optimization (PSO). However, linear decreasing inertia weight PSO (LDIW-PSO) algorithm is known to have the shortcoming of premature convergence in solving complex (multipeak) optimization problems due to lack of enough momentum for particles to do exploitation as the algorithm approaches its terminal point. Researchers have tried to address this shortcoming by modifying LDIW-PSO or proposing new PSO variants. Some of these variants have been claimed to outperform LDIW-PSO. The major goal of this paper is to experimentally establish the fact that LDIW-PSO is very much efficient if its parameters are properly set. First, an experiment was conducted to acquire a percentage value of the search space limits to compute the particle velocity limits in LDIW-PSO based on commonly used benchmark global optimization problems. Second, using the experimentally obtained values, five well-known benchmark optimization problems were used to show the outstanding performance of LDIW-PSO over some of its competitors which have in the past claimed superiority over it. Two other recent PSO variants with different inertia weight strategies were also compared with LDIW-PSO with the latter outperforming both in the simulation experiments conducted.

Stock Price Prediction Using the ARIMA Model

Stock price prediction is an important topic in finance and economics which has spurred the interest of researchers over the years to develop better predictive models. The autoregressive integrated moving average (ARIMA) models have been explored in literature for time series prediction. This paper presents extensive process of building stock price predictive model using the ARIMA model. Published stock data obtained from New York Stock Exchange (NYSE) and Nigeria Stock Exchange (NSE) are used with stock price predictive model developed. Results obtained revealed that the ARIMA model has a strong potential for short-term prediction and can compete favourably with existing techniques for stock price prediction.

Comparison Study of Swarm Intelligence Techniques for the Annual Crop Planning Problem

Annual crop planning (ACP) is an NP-hard type optimization problem in agricultural planning. It involves finding the optimal solution for the seasonal hectare allocations of a limited amount of agricultural land, among various competing crops that are required to be grown on it. This study investigates the effectiveness of employing three relatively new swarm intelligence (SI) metaheuristic techniques in determining the solutions to the ACP problem with case study from an existing irrigation scheme. The SI metaheuristics studied are cuckoo search (CS), firefly algorithm (FA), and glowworm swarm optimization (GSO). Solutions obtained from these techniques are compared with that of a similar population-based technique, namely, genetic algorithm (GA). Results obtained show that each of the three SI algorithms provides superior solutions for the case studied.

Classification of Phishing Email Using Random Forest Machine Learning Technique

Phishing is one of the major challenges faced by the world of e-commerce today. Thanks to phishing attacks, billions of dollars have been lost by many companies and individuals. In 2012, an online report put the loss due to phishing attack at about

A heuristic solution to the university timetabling problem

1.5 billion. This global impact of phishing attacks will continue to be on the increase and thus requires more efficient phishing detection techniques to curb the menace. This paper investigates and reports the use of random forest machine learning algorithm in classification of phishing attacks, with the major objective of developing an improved phishing email classifier with better prediction accuracy and fewer numbers of features. From a dataset consisting of 2000 phishing and ham emails, a set of prominent phishing email features (identified from the literature) were extracted and used by the machine learning algorithm with a resulting classification accuracy of 99.7% and low false negative (FN) and false positive (FP) rates.

A comparative study of some real-coded genetic algorithms for unconstrained global optimization

Purpose – The purpose of this paper is to consider the problem of university lecture timetabling. Timetabling deals with the problem of placing certain resources into a limited number of time slots, subject to given constraints, in order to satisfy a set of stated objectives to the highest possible extent. It is a well‐known and established NP‐hard problem. University timetabling is a major administrative activity especially in the third world universities. Solving the problem requires dynamic heuristics with predictable performance especially as the number of courses increases without corresponding increase in needed resources.Design/methodology/approach – A genetic algorithm metaheuristic is designed to handle a real‐life case study. Given the present structure of the case study, a modular approach to the design of the timetable schedules is adopted. The approach considers timetable in a bottom‐up fashion at the various levels of department, faculty or entire university. Simulation study is conducted us...

A multi-level genetic algorithm for a multi-stage space allocation problem

In this paper, a set of new real-coded genetic algorithms (RCGAs) with local and global exploratory search capabilities are proposed. The search capabilities are based on the inclusion of a modified crossover (MC) procedure and a new global exploratory method in RCGA. The global exploratory method is based on vector projection while the MC procedure is based on a limited version of the pattern search method. These modifications are introduced to increase the efficiency and robustness of RCGAs through better local and global exploration of the search region. An experimental study of the new algorithms was carried out using a set of 57 test problems. Statistical analyses and comparisons of the new algorithms with standard real-coded genetic algorithm (SRCGA) and some recent global optimization algorithms were carried out. Results obtained show that the modifications remarkably improve the performance of RCGAs across the test problems.

Simulated annealing based symbiotic organisms search optimization algorithm for traveling salesman problem

A new case of space allocation problem is considered. The study is based on a real-world multi-stage hostel space allocation for university students. A multi-level application of genetic algorithm metaheuristic with promising results is presented. Based on the case study, we examined the sensitivity analysis of various genetic algorithm operators in order to establish the baseline for practical deployment. The feasibility rate of the solutions obtained were also determined and presented.

On the performance of particle swarm optimisation without some control parameters for global optimisation

Symbiotic Organisms Search (SOS) algorithm is an effective new metaheuristic search algorithm, which has recently recorded wider application in solving complex optimization problems. SOS mimics the symbiotic relationship strategies adopted by organisms in the ecosystem for survival. This paper, presents a study on the application of SOS with Simulated Annealing (SA) to solve the well-known traveling salesman problems (TSPs). The TSP is known to be NP-hard, which consist of a set of (n1)!/2 feasible solutions. The intent of the proposed hybrid method is to evaluate the convergence behaviour and scalability of the symbiotic organisms search with simulated annealing to solve both small and large-scale travelling salesman problems. The implementation of the SA based SOS (SOS-SA) algorithm was done in the MATLAB environment. To inspect the performance of the proposed hybrid optimization method, experiments on the solution convergence, average execution time, and percentage deviations of both the best and average solutions to the best known solution were conducted. Similarly, in order to obtain unbiased and comprehensive comparisons, descriptive statistics such as mean, standard deviation, minimum, maximum and range were used to describe each of the algorithms, in the analysis section. The Friedmans Test (with post hoc tests) was further used to compare the significant difference in performance between SOS-SA and the other selected state-of-the-art algorithms. The performances of SOS-SA and SOS are evaluated on different sets of TSP benchmarks obtained from TSPLIB (a library containing samples of TSP instances). The empirical analysis results show that the quality of the final results as well as the convergence rate of the new algorithm in some cases produced even more superior solutions than the best known TSP benchmarked results.