Chorng-Shyong Ong

Building credit scoring models using genetic programming

Credit scoring models have been widely studied in the areas of statistics, machine learning, and artificial intelligence (AI). Many novel approaches such as artificial neural networks (ANNs), rough sets, or decision trees have been proposed to increase the accuracy of credit scoring models. Since an improvement in accuracy of a fraction of a percent might translate into significant savings, a more sophisticated model should be proposed to significantly improving the accuracy of the credit scoring mode. In this paper, genetic programming (GP) is used to build credit scoring models. Two numerical examples will be employed here to compare the error rate to other credit scoring models including the ANN, decision trees, rough sets, and logistic regression. On the basis of the results, we can conclude that GP can provide better performance than other models.

Two-stage genetic programming (2SGP) for the credit scoring model

Abstract Credit scoring models have been widely studied in the areas of statistics, machine learning, and artificial intelligence (AI). Many novel approaches such as artificial neural networks (ANNs), rough sets, or decision trees have been proposed to increase the accuracy of credit scoring models. Since an improvement in accuracy of a fraction of a percent might translate into significant savings, a more sophisticated model should be proposed for significantly improving the accuracy of the credit scoring models. In this paper, two-stage genetic programming (2SGP) is proposed to deal with the credit scoring problem by incorporating the advantages of the IF–THEN rules and the discriminant function. On the basis of the numerical results, we can conclude that 2SGP can provide the better accuracy than other models.

Model identification of ARIMA family using genetic algorithms

ARIMA is a popular method to analyze stationary univariate time series data. There are usually three main stages to build an ARIMA model, including model identification, model estimation and model checking, of which model identification is the most crucial stage in building ARIMA models. However there is no method suitable for both ARIMA and SARIMA that can overcome the problem of local optima. In this paper, we provide a genetic algorithms (GA) based model identification to overcome the problem of local optima, which is suitable for any ARIMA model. Three examples of times series data sets are used for testing the effectiveness of GA, together with a real case of DRAM price forecasting to illustrate an application in the semiconductor industry. The results show that the GA-based model identification method can present better solutions, and is suitable for any ARIMA models.

A novel hybrid model for portfolio selection

As we know, the performance of the mean-variance approach depends on the accurate forecast of the return rate. However, the conventional method (e.g. arithmetic mean or regression-based method) usually cannot obtain a satisfied solution especially under the small sample situation. In this paper, the proposed method which incorporates the grey and possibilistic regression models formulates the novel portfolio selection model. In order to solve the multi-objective quadric programming problem, multi-objective evolution algorithms (MOEA) is employed. A numerical example is also illustrated to show the procedures of the proposed method. On the basis of the numerical results, we can conclude that the proposed method can provide the more flexible and accurate results.

Variable selection in clustering for marketing segmentation using genetic algorithms

Marketing segmentation is widely used for targeting a smaller market and is useful for decision makers to reach all customers effectively with one basic marketing mix. Although clustering algorithms is popularly employed in dealing with this problem, it cannot be useful unless irrelevant variables are removed because irrelevant variables will distort the clustering structure and make the results useless. In this paper, genetic algorithms (GA) is used for variable selection and for determining the numbers of clusters. A real case of bank data set is used for illustrating the application of marketing segmentation. The results show that variable selection through GA can effectively find the global optimum solution, and the accuracy of the classified model is dramatically increased after clustering.

Optimal fuzzy multi-criteria expansion of competence sets using multi-objectives evolutionary algorithms

Competence set is widely used to plan the optimal expansion process of skills, abilities or strategies. However, the conventional method is concerned only with one criterion rather than multi-criteria problems. In addition, the crisp value cannot reflect the ambiguity and the uncertainty in practice. In this paper, we propose the fuzzy criteria competence set analysis. In order to obtain Pareto solutions, multi-objective evolutionary algorithm (MOEA) is employed here. A numerical example with two fuzzy criteria is also used to illustrate the proposed method.

A novel algorithm for uncertain portfolio selection

In this paper, the conventional mean-variance method is revised to determine the optimal portfolio selection under the uncertain situation. The possibilistic area of the return rate is first derived using the possibisitic regression model. Then, the Mellin transformation is employed to obtain the mean and the risk by considering the uncertainty. Next, the revised mean-variance model is proposed to deal with the problem of uncertain portfolio selection. In addition, a numerical example is used to demonstrate the proposed method. On the basis of the numerical results, we can conclude that the proposed method can provide the more flexible and accurate results than the conventional method under the uncertain portfolio selection situation.

Choosing best alliance partners and allocating optimal alliance resources using the fuzzy multi-objective dummy programming model

Synergy effects are the motives to enter into strategic alliances; however due to lack of adequate preparation or planning, these alliances often fail. It is of no doubt that a successful strategic alliance depends on choosing the correct alliance partners and appropriate resource allocation. In this paper, the fuzzy multi-objective dummy programming model is proposed to overcome the above-mentioned problems. Two types of strategic alliances, joint ventures and mergers and acquisitions (M&A), are demonstrated to choose the best alliance partners and allocate the optimal alliance resources in a numerical example. Based on the results, our method can provide the optimal alliance cluster and satisfaction in strategic alliances.

Interval multidimensional scaling for group decision using rough set concept

Abstract Multidimensional scaling (MDS) is a statistical tool for constructing a low-dimension configuration to represent the relationships among objects. In order to extend the conventional MDS analysis to consider the situation of uncertainty under group decision making, in this paper the interval-valued data is considered to represent the dissimilarity matrix in MDS and the rough sets concept is used for dealing with the problems of group decision making and uncertainty simultaneously. In addition, two numerical examples are used to demonstrate the proposed method in both the situation of individual differences scaling and the conventional MDS analysis with the interval-valued data, respectively. On the basis of the results, we can conclude that the proposed method is more suitable for the real-world problems.

Motivation and resource-allocation for strategic alliances through the DeNovo perspective

In recent years, there have been proposed many theories and models, such as transaction cost theory and the resource-based view, to explain the formation of strategic alliances. However, the perspectives are usually limited and incomplete. Additionally, the problem of resource allocation is also a serious issue when firms enter strategic alliances. This paper proposes a holistic perspective and provides an optimal resource portfolio by using the DeNovo perspective. A numerical example demonstrates the criteria of strategic alliances and provides the optimal resource portfolio.