• KIEE International Transactions on Power Engineering
• /
• 제5A권1호
• /
• pp.85-92
• /
• 2005

In this paper, Nash equilibriums of generation markets are investigated using a game theory application for simplified competitive electricity markets. We analyze the characteristics of equilibrium states in N-company spot markets modeled by uniform pricing auctions and propose a new method for obtaining Nash equilibriums of the auction. We assume that spot markets are operated as uniform pricing auctions and that each generation company submits its bids into the auction in the form of a seal-bid. Depending on the bids of generation companies, market demands are allocated to each company accordingly. The uniform pricing auction in this analysis can be formulated as a non-cooperative and static game in which generation companies correspond to players of the game. The coefficient of the bidding function of company-n is the strategy of player-n (company-n) and the payoff of player-n is defined as its profit from the uniform price auction. The solution of this game can be obtained using the concept of the non-cooperative equilibrium originating from the Nash idea. Based on the so called residual demand curve, we can derive the best response function of each generation company in the uniform pricing auction with N companies, analytically. Finally, we present an efficient means to obtain all the possible equilibrium set pairs and to examine their feasibilities as Nash equilibriums. A simple numerical example with three generation companies is demonstrated to illustrate the basic idea of the proposed methodology. From this, we can see the applicability of the proposed method to the real-world problem, even though further future analysis is required.

• 한국정보통신학회논문지
• /
• 제21권7호
• /
• pp.1299-1305
• /
• 2017

본 논문에서는 다중 안테나를 사용하는 두 개의 송신단이 각각 하나의 안테나를 가지고 있는 수신단을 서비스 할때 구성되는 두 유저 간섭 채널에서 불완전한 채널 정보를 사용하는 송신단들의 빔포밍 게임을 고려한다. 빔포밍 게임에서는 송신단(i.e., 플레이어)이 자신들의 전송률(i.e., 보상)을 높이기 위해서 경쟁을 하고, 빔포밍 기법을 위해서는 완전한 채널 상황에서 최적이라고 알려진 maximum ratio transmission (MRT) 기법과 zero forcing (ZF) 빔포밍 기법을 선형으로 결합하는 기법(i.e., 전략)을 사용한다. 우리가 제안하는 게임에서는 송신단이 불완전한 채널 정보를 사용하므로, 최적의 전략을 찾더라도 그 전략이 유효하지 않을 수 있다. 본 논문에서는 불완전한 채널 정보로부터 양자화 오차의 영향을 고려한 송신단의 효율적인 빔포밍 전략을 제안한다.

• 한국경영과학회지
• /
• 제34권1호
• /
• pp.1-10
• /
• 2009

Consumers earn a number of points for every purchase and then they can exchange a specified number of points for a desired reward in a typical loyalty program. The immediate payoff of their effort given as points is not the real reward they actually care about. It is merely an instrument (or medium) which has no value in itself. In a real world, consumers frequently choose the option with a bigger medium even though the economic value of the option is not changed by the medium. We call it 'medium effect.' In this study we explored if the size of medium affects consumers' preferences. For this we controlled the reward options with three types of medium (small, medium, big) and measured the magnitude of preference difference among the three types of reward options. In addition, we manipulated comparability of reward options with wine and gas discount coupon. We confirmed that choosing one of two wines was easier than that of the two gas coupons. 123 respondents were allocated into three experimental groups. In three experimental grounds, the ratios of the focused reward option's medium to the compared reward option's medium were different. For example, the focused reward option has 10 million points whereas the compared reward option has 10 million points for 1 million won purchase amount in the first group. Then each respondent was asked to choose one of two loyalty programs (focused program vs. compared program) in two different conditions (comparability between reward options. easiness vs. difficulty). To compare the medium effects among the experiment conditions we used chi-squares tests. The empirical results show consumer preference increases and then decreases as reward mile-ages/points given according to purchase amount increase. Additionally, they let us know that comparability of alter natives affects change of consumer preference by reward mileages/points.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2008년도 추계학술대회 및 정기총회
• /
• pp.179-190
• /
• 2008

Consumers earn a number of points for every purchase and then they can exchange a specified number of points for a desired reward in a typical loyalty program. The immediate payoff of their effort given as points is not the real reward they actually care about. It is merely an instrument (or medium) which has no value in itself. In a real world, consumers frequently choose the option with a bigger medium even though the economic value of the option is not changed by the medium. We call it 'medium effect'. In this study we explored if the size of medium affects on consumers' preferences. For this we controlled the reward options with three types of medium (small, medium, big) and measured the magnitude of preference difference among the three types of reward options. In addition, we manipulated comparability of reward options with wine and gas discount coupon. We confirmed that choosing one of two wines was easier than that of the two gas coupons. 164 respondents were allocated into three experimental groups and one control group. In three experimental groups, the ratios of the focused reward option's medium to the compared reward option's medium were different. For example, the focused reward option has 10 million points whereas the compared reward option has 10 million points for 1 million won purchase amount in the first group. Then each respondent was asked to choose one of two loyalty programs (focused program vs. compared program) in two different conditions (comparability between reward options: easiness vs. difficulty). To compare the medium effects among the experiment conditions we used chi-squares tests. The empirical results show the medium effect increases and then decreases as the ratio of reward mileages/points to purchase amount increases. Additionally, they let us know that comparability of alternatives affects on the medium effects depending on the ratio of reward mileages/points to purchase amount.

• Communications for Statistical Applications and Methods
• /
• 제29권1호
• /
• pp.85-101
• /
• 2022

Derivative-linked securities (DLS) is a type of derivatives that offer an agreed return when the underlying asset price moves within a specified range by the maturity date. The underlying assets of DLS are diverse such as interest rates, exchange rates, crude oil, or gold. A German 10-year bond rate-linked DLS and a USD-GBP CMS rate-linked DLS have recently become a social issue in Korea due to a huge loss to investors. In this regard, this paper accounts for the payoff structure of these products and evaluates their prices and fair coupon rates as well as risk measures such as Value-at-Risk (VaR) and Tail-Value-at-Risk (TVaR). We would like to examine how risky these products were and whether or not their coupon rates were appropriate. We use Hull-White Model as the stochastic model for the underlying assets and Monte Carlo (MC) methods to obtain numerical results. The no-arbitrage prices of the German 10-year bond rate-linked DLS and the USD-GBP CMS rate-linked DLS at the center of the social issue turned out to be 0.9662% and 0.9355% of the original investment, respectively. Considering that Korea government bond rate for 2018 is about 2%, these values are quite low. The fair coupon rates that make the prices of DLS equal to the original investment are computed as 4.76% for the German 10-year bond rate-linked DLS and 7% for the USD-GBP CMS rate-linked DLS. Their actual coupon rates were 1.4% and 3.5%. The 95% VaR and TVaR of the loss for German 10-year bond rate-linked DLS are 37.30% and 64.45%, and those of the loss for USD-GBP CMS rate-linked DLS are 73.98% and 87.43% of the initial investment. Summing up the numerical results obtained, we could see that the DLS products of our interest were indeed quite unfavorable to individual investors.

• Journal of Korea Trade
• /
• 제23권5호
• /
• pp.118-128
• /
• 2019

Purpose - This paper examines the international R&D contest in which the extent of intellectual property right (IPR) affects both the size of prize for the winning firm and the extent of positive spillover through cost of firms. Recognizing the possibility of incomplete protection of IPR, the present paper analyzes the effect of changes in the extent of IPR on payoffs to firms and social welfare. Design/methodology - This paper examines coordination of IPRs by countries in economic integration. The paper then develops a general model of international R&D contest with incomplete protection of IPR. An increase in the extent of IPR augments the share of the prize the winning firm can appropriate, while decreasing the positive cost externality. To derive sharper results, the paper considers the cases of linear and fixed spillovers. Findings - Under plausible assumptions, an increase in the IPR augments the payoff to each firm and the aggregate payoffs as well. The paper also shows that the number of firms participating in the R&D contest can be endogenously determined in the two-stage R&D contest. The higher the extent of cost spillover, and the larger the effective prize, the more firms participate in the international R&D contest. Originality/value - Existing studies assume that firms winning the R&D contest enjoy perfect IPR to the output of their R&D activities. This is a very restrictive assumption in that other firms can copy the new products or processes. By allowing for the incompleteness of the IPR, the present paper develops a more realistic model of R&D contest. The novelty of the present paper is to allow for the possibility that the higher extent of IPR increases the prize and decreases positive cost externality at the same time. The findings of the present paper can serve as a basis for government policy toward R&D activities of firms and protection of IPRs.

• 한국통신학회논문지
• /
• 제39B권11호
• /
• pp.790-802
• /
• 2014

인지무선 기술에서 주사용자의 보호를 위해 부사용자들은 주기적인 센싱 수행을 통해 주사용자의 부재를 판단하게 되고, 부사용자들 간의 협력 센싱을 통해서 향상된 센싱 결과를 얻을 수 있다. 하지만 주사용자에 대한 검출 확률과 오경보 확률에 대한 비용의 트레이드 오프가 존재하기 때문에, 적절한 협력 집단의 규모 유지가 필요하다. 또한 부사용자들은 자신이 현재 사용중인 주파수 대역은 물론 인가 사용자가 나타났을 시에 스위칭 해야 할 후보 채널에 대한 주기적인 센싱이 요구된다. 본 논문에서는 진화게임이론을 이용하여 분산상황 에서의 인밴드 센싱과 아웃밴드 센싱을 고려한 효율적인 그룹 협력 센싱 방법을 제안한다. 진화 게임을 통해서 협력센싱의 전략을 택한 부사용자들의 집단이 ESS(Evolutionary Stable State)상태로 수렴함을 관찰하였고, 학습 알고리즘을 통해 서로간의 정보교환 없이 평형상태로 수렴함을 관찰하였다.

• 농업과학연구
• /
• 제29권1호
• /
• pp.55-65
• /
• 2002

The most representative fruit in Yanbian area is Applepear(Ping Guo Li). It has been a important resource of farm income. According to the economic growth and income Increase of individual consumers, food consumption pattern will be changed from grain to high value cash crops such as Applepear and vegetables. The globalization and adapt ion of free market oriented economic policy of China government have led to change the collective farming system to individual farm management system. The institutional transformation have brought about high productivity of farm products and incomes of farmers. Therefore the plantation area of Applepear and requirement of investment cost for establishment of Applepear farms have increased rapidly since the 1980's, the time of perestroika and glasnost in Chinese economy. In prosperity of Chinese agriculture, individual farmer's decision making as free selection of farmland use, selection of high pay-off crops, free sale of the products in free market and credit support for the project might be the most important factors. In case of Applepear farm development, net present value of the project net benefit was estimated at 55.518won per hectare and financial rate of return was 21%. The benefit/cost ratio of the project was 2.11. Considering 10% of discounting rate or the opportunity cost of capital in China the Applepear farm development project showed us economically feasible in the light of the above efficiency indices. The Chinese government has to support Applepear farm development project financially and institutionally considering the high-payoff benefit of the fruit and farmer's in come increase in the future.

• 한국통신학회논문지
• /
• 제38B권7호
• /
• pp.543-552
• /
• 2013

차량통신간 자동차의 정보를 포함하는 메시지인 비콘을 주변의 자동차들에게 주기적으로 전송함으로 차량간의 안전을 보장해야 할 필요가 있다. 그러나 IEEE 802.11p/WAVE 환경에서는 혼잡상황에서 비콘 전송을 경쟁윈도우(Contention window)를 기반으로 한 MAC (Medium Access Control)에서 해결하도록 하였기 때문에 비콘 전송 과정에 충돌이 발생한다. 본 논문에서는 차량 혼잡 상황에서 비콘의 충돌을 막기 위해서 MAC 구조를 슬롯 p-persistent CSMA (Carrier Sense Multiple Access)라 가정하여 전송 확률에 대한 보수행렬(payoff)을 이용한 비 협조적 베이지안 게임이론을 적용하여 베이지안 내쉬 균형점(BNE)을 도출하였다. 본 논문에서는 도출된 BNE를 가지고 포화 상태에서 비콘 전송률을 높일 수 있는 혼잡 제어 알고리즘을 제안했다. 제안된 알고리즘으로 경쟁윈도우크기와 주변 자동차 수에 관련된 함수를 통해 전송 확률을 계산하였다. 시뮬레이션 결과를 통해서 제안된 알고리즘이 안정적으로 동작함을 보였다.

• 경영과정보연구
• /
• 제33권1호
• /
• pp.45-63
• /
• 2014

주식시장에서 트레이딩 시스템(trading system)은 투자 위험을 줄이면서 수익을 극대화하기 위해 일정한 규칙을 컴퓨터에 적용하여 기계적으로 거래를 실행하는 자동매매시스템이다. 그러나 단일 자산을 거래 대상으로 하는 트레이딩 시스템 개발은 그 자산의 움직임에 의해 수익 구조가 결정되는 문제점을 가지고 있기 때문에 위험을 초래할 우려가 존재한다. 따라서 두 자산 간 상대적 가치를 활용하여 한 자산의 위험 발생 시 다른 자산의 거래를 통해 안정적인 수익을 창출할 수 있는 트레이딩 시스템 개발에 대해 주목할 필요가 있다. 본 연구에서는 두 개의 주가지수 선물 간의 상대적 가치를 이용한 규칙기반 트레이딩 시스템을 개발하는 것이 목적이다. 이를 위해 차익거래 전략 시 활용되는 두 자산의 가격비율을 기반으로 기술적 지표(technical indicator)들을 생성하고, 데이터 마이닝 기법인 러프 집합 분석을 이용하여 거래규칙을 생성하였다. 생성된 거래규칙의 유용성을 확인하기 위해 슬라이딩 윈도우(sliding window) 방법을 사용하여 거래 시뮬레이션을 적용하였으며, KOSPI 200 지수선물과 S&P 500 지수선물을 분석 대상으로 선정하여 성과를 측정하였다. 실험 결과, 제안한 시스템이 시장의 흐름과 상관없이 절대적인 수익을 발생하는 것을 확인 할 수 있었다.