Napat Harnpornchai

Portfolio optimization using multi-obj ective genetic algorithms

A portfolio optimisation problem involves allocation of investment to a number of different assets to maximize yield and minimize risk in a given investment period. The selected assets in a portfolio not only collectively contribute to its yield but also interactively define its risk as usually measured by a portfolio variance. In this paper we apply various techniques of multiobjective genetic algorithms to solve portfolio optimization with some realistic constraints, namely cardinality constraints, floor constraints and round-lot constraints. The algorithms experimented in this paper are Vector Evaluated Genetic Algorithm (VEGA), Fuzzy VEGA, Multiobjective Optimization Genetic Algorithm (MOGA) , Strength Pareto Evolutionary Algorithm 2nd version (SPEA2) and Non-Dominated Sorting Genetic Algorithm 2nd version (NSGA2). The results show that using fuzzy logic to combine optimization objectives of VEGA (in VEGAFuzl) for this problem does improve performances measured by Generation Distance (GD) defined by average distances of the last generation of population to the nearest members of the true Pareto front but its solutions tend to cluster around a few points. MOGA and SPEA2 use some diversification algorithms and they perform better in terms of finding diverse solutions around Pareto front. SPEA2 performs the best even for comparatively small number of generations. NSGA2 performs closed to that of SPEA2 in GD but poor in distribution.

Risk analysis of Thalassemia using knowledge representation model: Diagnostic Bayesian Networks

Bayesian Networks (BNs) is one of the most effective theoretical models applied to make medical diagnostic decisions. In particular, it has been applied to Thalassemia, which is one of the most common genetic disorders in the world. The main problems of diagnosing this disease are the complex processes for diagnosing the several types of Thalassemia which occur in Thailand. Moreover, diagnostic methods are slow and rely on expert knowledge and experience as well as expensive equipment. The advantage of BNs is that they are used to represent the diagnostic domain in the form of graphical statistical models. The propose of this paper is to construct a Diagnostic Bayesian Networks for risk analysis of Thalassemia using polychromatic model for screening each type of Thalassemia, including related variables. The model will be used to elicit and calculate the probabilities of each type of Thalassemia in future research.

Dynamic adjustment of age distribution in Human Resource Management by genetic algorithms

Adjustment of a given age distribution to a desired age distribution within a required time frame is dynamically performed for the purpose of human resource (HR) planning in human resource management (HRM). The adjustment process is carried out by adding the adjustment magnitudes to the existing number of employees at the selected age groups on the yearly basis. A model of a discrete dynamical system is employed to emulate the evolution of the age distribution used under the adjustment process. Genetic algorithms (GA) is applied for determining the adjustment magnitudes that influence the dynamics of the system. An interesting aspect of the problem lies in the high number of constraints; though the constraints are fundamental, they are considerably higher in number than in many other optimization problems. An adaptive penalty scheme is proposed for handling the constraints. Numerical examples show that GA with the utilized adaptive penalty scheme provides potential means for HR planning in HRM.

A Flood Risk Assessment Based on Maximum Flow Capacity of Canal System

System analysis of network of flows of water is essential for assessing risk of flooding. The flood risk management generally focuses on the meteorological forecasting together with the operation of hydraulic structure while overlooks the flood incurred by inefficient performance of natural instrument in flood mitigation, namely of canal systems. A new methodology for the risk assessment of flood from the prospect of the capacity of canal system is proposed in this paper. The methodology comprises the modeling of a canal system by a flow network in the graph theory, the formulation for the determination of the system capacity in terms of the maximum flow problem, the treatment of uncertainty using copula couple with maximum entropy models, the definition of flood risk event, and the method of risk assessment. The application of the proposed methodology is illustrated through a numerical example.

Handling Model Risk in Portfolio Selection Using Multi-Objective Genetic Algorithm

The application of multi-objective optimization to portfolio optimization has attracted increased attention in recent years. Portfolio optimization is defined as the task to select a combination of assets that simultaneously satisfy two objectives, namely maximized portfolio return and minimized portfolio volatility of return. In its original application, Markowitz (Journal of Finance, 7(1):77–91, 1952) used mean returns and the variances from each of those asset’s returns as inputs for the model. However, the ex-post results of portfolio optimization based on these historical assets’ mean returns and volatilities are sub-optimal. Although the historical mean and volatilities are substituted by forecasted returns and volatilities, the ex-post results are still sub-optimal due to the inaccuracy of the forecasting models leading to what is known as model risk.

Analysis of branching ratio of telecommunication stocks in Thailand using hawkes process

The aim of the research is to study the branching ratios of the telecommunication stocks in Thailand, ADVANC and DTAC, both listed on the Stock Exchange of Thailand (SET). The branching ratio is the parameter defined in the Hawkes process and directly measures the influential degree of endogeneity. The results indicate to what extent the stock price changes are affected by internal factors. The study found that the branching ratio of ADVANC is at 29 %, which means ADVANCs price change is only 29 %, caused by internal factors, while the remaining 71 % derives from external factors. Meanwhile, DTACs branching ratio is at 55 %, meaning DTACs price change is 55 % due to internal factors and 45 % due to external. Knowing to what extent the stock price is affected by external factors can strengthen investor strategy. Stocks with a low branching ratio are more speculative than those having a high branching ratio.

Forecasting stock returns using variable selections with Genetic Algorithm and Artificial Neural-Networks

Modeling stock returns requires selections of appropriate input variables. For an Artificial Neural Network, the appropriate input variables have both linear and nonlinear functional relationship with stock returns as output variables. To capture the non-linear relationships, we propose Weierstrass theorem. However, to estimate the relationships for all possible combinations of input variables, especially for a large set of variables, is too numerous for a simple exhaustive search thus we use a Genetic Algorithm to approximate the non-linear relationships between the prospective input variables and the output variables. The result shows that the Artificial Neural Networks with the selected variables based on both linear and non-linear relationship perform better than the ones with all possible variables for all but one out of the sample of ten US stocks.