Optimization of Support Vector Machines for Financial Forecasting (재무예측을 위한 Support Vector Machine의 최적화)
-
- Journal of Intelligence and Information Systems
- /
- v.17 no.4
- /
- pp.241-254
- /
- 2011
Financial time-series forecasting is one of the most important issues because it is essential for the risk management of financial institutions. Therefore, researchers have tried to forecast financial time-series using various data mining techniques such as regression, artificial neural networks, decision trees, k-nearest neighbor etc. Recently, support vector machines (SVMs) are popularly applied to this research area because they have advantages that they don't require huge training data and have low possibility of overfitting. However, a user must determine several design factors by heuristics in order to use SVM. For example, the selection of appropriate kernel function and its parameters and proper feature subset selection are major design factors of SVM. Other than these factors, the proper selection of instance subset may also improve the forecasting performance of SVM by eliminating irrelevant and distorting training instances. Nonetheless, there have been few studies that have applied instance selection to SVM, especially in the domain of stock market prediction. Instance selection tries to choose proper instance subsets from original training data. It may be considered as a method of knowledge refinement and it maintains the instance-base. This study proposes the novel instance selection algorithm for SVMs. The proposed technique in this study uses genetic algorithm (GA) to optimize instance selection process with parameter optimization simultaneously. We call the model as ISVM (SVM with Instance selection) in this study. Experiments on stock market data are implemented using ISVM. In this study, the GA searches for optimal or near-optimal values of kernel parameters and relevant instances for SVMs. This study needs two sets of parameters in chromosomes in GA setting : The codes for kernel parameters and for instance selection. For the controlling parameters of the GA search, the population size is set at 50 organisms and the value of the crossover rate is set at 0.7 while the mutation rate is 0.1. As the stopping condition, 50 generations are permitted. The application data used in this study consists of technical indicators and the direction of change in the daily Korea stock price index (KOSPI). The total number of samples is 2218 trading days. We separate the whole data into three subsets as training, test, hold-out data set. The number of data in each subset is 1056, 581, 581 respectively. This study compares ISVM to several comparative models including logistic regression (logit), backpropagation neural networks (ANN), nearest neighbor (1-NN), conventional SVM (SVM) and SVM with the optimized parameters (PSVM). In especial, PSVM uses optimized kernel parameters by the genetic algorithm. The experimental results show that ISVM outperforms 1-NN by 15.32%, ANN by 6.89%, Logit and SVM by 5.34%, and PSVM by 4.82% for the holdout data. For ISVM, only 556 data from 1056 original training data are used to produce the result. In addition, the two-sample test for proportions is used to examine whether ISVM significantly outperforms other comparative models. The results indicate that ISVM outperforms ANN and 1-NN at the 1% statistical significance level. In addition, ISVM performs better than Logit, SVM and PSVM at the 5% statistical significance level.
This study is designed to examine feasibility and limitation of outsourcing in cooking Western food in a hotel, to interpret importance of outsourcing(eg. outside order, outside procurement, outside supply) in a broad sense in order to reinforce the core capacity in the cooking department, and to know whether the cooking human power is efficiently used and how much the chefs recognize outsourcing of food materials. As many companies conduct restructuring to cut down its size, the reduction of human power led the Western food cooking in the hotel to lower core capacities, lower quality, and lower efficiency. In addition, the sagging morale of chefs undermined creativity. To change from the traditional kitchen to an economic kitchen needs to look into importance of outsourcing, cognitive attitude of chefs, relation with outside suppliers. Here suggests performance of positive changes in the structure The study examined feasibility and limitation of outsourcing in the hotel kitchen as well as chefs' cognitive attitude toward outsourcing of food materials to reinforce core capabilities of the hotel kitchen. 1. Companies of outsourcing are selected according to variability of price conditions, flexibility of contract conditions, popularity of the outsourcing company, and reputation of the outsourcing company. 2. The importance of outsourcing in the Western food cooking is divided into 4 factors such as standard of selecting outsourcing companies, policies of cooking manu, quality of cooking, and quantity of cooking. 3. The most feasible section in outsourcing of food materials is a process of kneading flour for bread, which shows that many Western-food chefs expect to put higher possibility of outsourcing on the kneading. In other words, when it comes to confectionery and bakery, there are many outside expert processing companies supplying high quality products. In the order of outsourcing feasibility, sauce is followed by processed vegetable, garnish of main dish, and soup. The least feasible section in outsourcing of food materials is appetize. Appetize includes a concept of a improvised dish and needs speed. Due to its color, freshness, and sensibility of taste, the appetize plays a key role in the Western food cooking. 4. When outsourcing is taken in place, the highest risk is to lower the inner cooking skills. Therefore chefs in charge of the Western food sequently recognize both internal problems including storage of cooking skills, unstability of layoffs, and loss of cooperation between departments, and external problems including inferior goods, difficulty of differentiating manu, delay of delivery, and expiration date. It shows that most of the Western food chefs consider risks of the internal problems at first. 5. A effective outsourcing needs appropriate selection of outsourcing companies, maintenance of credibility, active communication, check and management of hygiene. However regardless of their position or career, chefs in charge of the Western food have the same cognitive attitude toward selecting successful outsourcing companies after the outsourcing system is enforced. The core of cooking, or a final stage in the full process of so-called artistic cooking, should be treated with insourcing. Reduction of several cooking processes resulted in shortened cooking time, increased efficiency, faster cooking, cutting the waiting-lines, and finally more room for customers. The outsourcing system can reduce or eliminate the following processes in cooking: buying various food materials, checking, storing, preparing, and processing. Especially in the Western food cooking department of a hotel, the outsourcing system should be enforced to make an economic kitchen and to efficiently manage it. Wow it's time to change from the traditional kitchen to an economic kitchen in the hotel cooking department. For that, the cooking department should become a small but strong organization by outsourcing except its core work.
Franchising is one of the fastest growing types of business. It is already popular and well-known in the U.S., and has been growing in many other countries including Korea. Furthermore, many Korean franchising companies have expanded their business overseas actively. According to the data by the Ministry of Industry and Resource, 82 companies out of a sample of 500 franchising companies are already operating in many foreign countries and 48% of them have started their foreign business since 2006. This clearly indicates the fast growing current trend of foreign operation by Korean franchising companies. In spite of the fast growing trend of foreign expansion in the industry, academic research on internationalization of franchising companies is extremely difficult to find. Accordingly, academic research on the issue is necessary and urgent in Korea. Among the various research questions on internationalization of franchising business, this study intends to investigate the difference in organizational factors between the franchising companies doing foreign operation and those doing business only domestically. More specifically, this research has the following purposes. First, considering the lack of theoretical basis of previous studies, resource-based theory and agency theory are employed as the theoretical bases. Second, this study explains the difference in internationalization based on organizational factors such as company size, history and growth rate. Third, the five hypotheses regarding the difference in organizational factors are presented and tested empirically, which is the first attempt in the area of this topic. Finally, the study attempts to clarify the conflicting implications among theories regarding some organizational factos such as growth rate. As the theoretical background, resource-based theory and agency theory are discussed. According to resource-based theory, a firm can grow continuously when it has competence and resource, and also the ability to develop them. The competence and resource can include capital, human resource, management skill, market information, ability to manage risk, etc. Meanwhile, agency theory views the relationship between franchisor and franchisee as an agency relationship. In agency theory, bonding capability and monitoring capability are the two key factors which promote internationalization of franchising companies. Based on the two theories, a conceptual model is designed. The model consists of two groups of variables. One is organizational factors including size, history, growth rate, price bonding and geographic dispersion. The other is whether a franchising company is operating overseas or not. We developed the following five research hypotheses basically describing the relationship between organizational factors and internationalization of franchising companies. H1: The size of franchising companies operating overseas is larger than that of franchising companies operating domestically. H2: The history of franchising companies operating overseas is longer than that of franchising companies operating domestically. H3: The growth rate of franchising companies operating overseas is higher than that of franchising companies operating domestically. H4: The price bonding of franchising companies operating overseas is higher than that of franchising companies operating domestically. H5: The geographic dispersion of franchising companies operating overseas is wider than that of franchising companies operating domestically. Data for the analyses are obtained from 2005 Korea Franchise Survey data co-generated by Ministry of Industry and Resource, GS1 Korea, and Korea Franchise Association. Out of 2,804 population companies, 2,489 companies are excluded for various reasons and 315 companies are selected as the final sample. Prior to hypotheses tests, validity and reliability of the measures of size, history, growth rate and price bonding are examined for further analyses. Geographic dispersion is not validated since it is measured using nominal data. A series of independent sample T-tests is used to find out whether there exists any significant difference between the companies internationalized and those operating only domestically for each organizational factor. Among the five factors, size and geographic dispersion show significant difference, growth rate and price bonding do not reveal any difference and, finally, history factor shows conflicting results in the difference depending on how to measure it.