• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.3
• /
• pp.63-72
• /
• 2020

The study aims to investigate the impact level of earnings quality on firm value. The study has used data with 3,910 observations at listed firms on Vietnam Stock Exchange for the period from 2010 to 2018, and GLS regression analysis is employed in this research. Earnings quality is measured in the aspects of earnings management, earnings persistence, and timeliness of profitability. This study also considers a number of controlled variables that positively influence the firm's value such as firm size, fixed asset investment rate and dividend payout ratio. The results show that earnings quality is positively associated with firm value with having statistical significance. In contrast, some determinants negatively influence firm value such as financial leverage, ratio of market value to book value, and revenue growth. Determinants of firm size, the rate of investment in fixed assets, the rate of dividend payment positively affect the firm value. In contrast, determinants of financial leverage, revenue growth rate and market value to book value ratio are inversely related to firm value according to economic value, Tobin's Q or Price. Based on the findings, some recommendations are proposed for investors, management and policy makers as well in the context of emerging countries including Vietnam.

• The Journal of the Convergence on Culture Technology
• /
• v.2 no.4
• /
• pp.49-54
• /
• 2016

Development of AR/VR has expanded computer games from indoor static contents to outdoor dynamic contents. The syndrome of 'Pokemon Go' which is a representative AR game was born, and Nintendo's stock price shot up a massive 120%. Although some people gained profits, new damages are emerging including illegal intrusion or road accidents because it is a type of game to visit real locations. There are reported about 114 thousand road accidents related to the game. This study aims to examine the trends of AR game in 2016 and analyze damages and positive cases to suggest a method for preventing accidents due to the AR game, and a business model.

• The Korean Journal of Applied Statistics
• /
• v.29 no.1
• /
• pp.171-180
• /
• 2016

Basel committee suggests using Value-at-Risk (VaR) and expected shortfall (ES) as a measurement for market risk. Various estimation methods of VaR and ES have been studied in the literature. This paper compares semi-parametric methods, such as conditional autoregressive value at risk (CAViaR) and conditional autoregressive expectile (CARE) methods, and a Gaussian quasi-maximum likelihood estimator (QMLE)-based method through back-testing methods. We use unconditional coverage (UC) and conditional coverage (CC) tests for VaR, and a bootstrap test for ES to check the adequacy. A real data analysis is conducted for S&P 500 index and Hyundai Motor Co. stock price index data sets.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.11
• /
• pp.769-778
• /
• 2020

The purpose of this study was to examine the role of profitability as a mediating variable in influencing firm value. This study uses a sample of manufacturing companies listed on the Indonesia Stock Exchange from 2016 to 2018. The data used is panel data, with data analysis using multiple regression. Based on the Sobel test, profitability plays a role in mediating the effect of firm size on firm value. The effect of firm size on firm value is indirect, however, through profitability. Therefore, the market price of the shares of large-scale companies will increase if the resulting profitability is high. The capital structure and managerial ownership directly influence firm value. The results showed that managerial ownership and firm size had a positive effect on profitability, while capital structure had no effect on profitability. Capital structure and managerial ownership have a negative effect on firm value, while firm size and profitability have a positive effect on firm value. The main finding of this study is that profitability acts as an intervening variable in mediating the relationship between firm size and firm value.

• East Asian Economic Review
• /
• v.20 no.4
• /
• pp.469-491
• /
• 2016

In this paper, we estimate the exchange rate exposure, indicating the effect of exchange rate movements on firm values, for a sample of 1,400 firms in seven East Asian countries. The exposure estimates based on various exchange rate variables, return horizons and a control variable are compared. A key result from our analysis is that the long term effect of exchange rate movements on firm values is greater than the short term effect. And we find very similar results from using other exchange rate variables such as the U.S. dollar exchange rate, etc. Second, we add exchange rate volatility as a control variable and find that the extent of exposure is not much changed. Third, we examine the changes in exposure to exchange rate volatility with an increase in return horizon. Consequently the ratio of firms with significant exposures increases with the return horizons. Interestingly, the increase of exposure with the return horizons is faster for exposure to volatility than for exposure to exchange rate itself. Taken as a whole, our findings suggest that the socalled "exposure puzzle" may be a matter of the methodology used to measure exposure.

• Asian Journal of Business Environment
• /
• v.11 no.4
• /
• pp.31-38
• /
• 2021

Purpose: This paper uses the resource-dependency theory to present the case of the Pakistan sugar industry to highlight how the industry uses a strategic alliance to gain a powerful bargaining position over its critical dependencies. The case of the Pakistan sugar industry is well-known and it is common knowledge that the alliance or the cartel within it is responsible for frequent price hikes and sugar supply shortages in the country. Research design, data and methodology: We use a case study, qualitative document analysis design to trace how the alliance overcomes its various dependencies, and in doing so, how does it harm various stakeholder interests. Results: This paper finds that the sugar industry alliance maintains its bargaining power by manipulating sugar supply through horizontal alliances, political affiliations, underselling and under-reporting sugar stocks, purchasing sugarcane from the black market, and by gaining billions of rupees in export subsidies by hoarding stock and using its political connections. Conclusion: The paper concludes by providing a summary of the measures which the government has taken to curb this anticompetitive conduct; the most important of which is the removal of protectionist measures for sugar trade and allowing market forces to control the demand and supply of sugar in the local market.

• Asia-Pacific Journal of Business
• /
• v.13 no.1
• /
• pp.247-262
• /
• 2022

Purpose - The purpose of this study is to examine the impact of the Pandamic Declaration on 'untact firms' listed in KOSPI and KOSDAQ market in order to verify Investor Mania Hypothesis. Design/methodology/approach - This study collected financial data for 44 untact firms in KOSPI and KOSDAQ market. Then, we employed ESM(Event Study Methodology), EGARCH model and DID(Difference-In-Difference) for analysis. Findings - First, in contrast with the benchmarking index, KOSPI 200 which shows a negative (-) abnormal return trend, the untact firms have positive abnormal return trend consistently. Second, after the Pandemic Declaration, the variability of abnormal return for the untact firms is found to be significantly positive. Third, we find that the cumulative abnormal return and volatility of the untact firms significantly increase after the Pandemic Declaration. Research implications or Originality - Based on the Investor Mania Hypothesis, we confirm that the market potential of untact firms after the Pandemic Declaration is observed when compared with the KOSPI 200.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.24 no.6
• /
• pp.873-883
• /
• 2000

The purpose of this study were to investigate the relationships of situation variables, product variables. consumer variables and impulse buying behavior in apparel store. We collected data from 462 consumers of adult women living in Pusan and analysed by factor analysis, cluster analysis, analysis of variance, t-test and discriminant analysis. The results were as follows: First, The purchase situation influenced on the impulse buying behavior consisted of the Pre-purchase condition and the Point-of-purchase state. The in-store situation consisted of the Salesman/store atmosphere, the Low price and the Possibility of out of stock. And the consumption values of apparel are divided into four factors ; Emotional/aesthetical value, Epistemic value, Functional value and Social value. The clothing shopping orientation as consumer variable extracted six factors ; Recreational orientation, Economical orientation, Brand/store loyalty orientation, Careful orientation, Apathetic orientation and Positive orientation. Consumers were classified by the cloting shopping orientation into the Convenience shopper, the Recreational shopper, the Economical shopper and the Careful shopper. Second, In comparison with the unimpulse-buyin groups, the impulse-buying group is more effected by in-store situation than purchase situation, and were more effected by Emotional/aesthetical value, Social value and Epistemic value of the consumption value. In consumer types, the more was the Recreational shopper and the Convenience shopper, the more showed impulse buying behavior. And the important factor distinguished between the impulse buying group and the unimpulse buying group was the Salesman/store atmosphere of the in-store situation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.10
• /
• pp.4354-4358
• /
• 2011

Many Korean construction companies have fallen on hard times because the construction business continues to stagnate. Therefore it is necessary to measure the management efficiency for efficient operation and strengthening competitiveness of them in order to survive a difficult situation. This paper proposes a combined ANP and DEA model to analyze the efficiency of the listed construction firms. In order to determine the input and output variables of DEA, the ANP model is applied to evaluate the importance of input and output variables. The benchmarking companies and efficiency value for the construction firms with inefficiency are also provided to improve the their efficiency. The 57 listed construction companies consisted of 36 listed on KOSPI and 21 listed on KOSDAQ are analyzed in this study. The analysis results show that 11 companies whose values of CCR are 1, and 14 firms whose values of BCC efficiency are 1. In additions, the 19 firms have the scalability efficiency. Finally, we test the correlation between efficiency and the stock price.