선택실험법 자료에서의 선호이질성 분석을 위한 혼합로짓모형 및 잠재계층모형의 활용

유병국 ( Byong Kook Yoo ) 한국환경경제학회·한국자원경제학회(구 한국환경경제학회) 2012 자원·환경경제연구 Vol.21 No.4

조건부 로짓(Conditional Logit: CL)모형은 모형추정 및 결과해석이 비교적 용이하다는 장점으로 널리 사용되는 반면에 응답자의 선호이질성(preference heterogeneity)을 충분히 반영하지 못한다는 한계를 가지고 있다. 본 연구에서는 최근 지배적인 방식으로 제시되고 있는 두개의 모형인 혼합로짓(Mixed Logit; ML)모형과 잠재계층모형(Latent Class Model; LCM)을 사용하여 우리나라 보령댐 주변 습지보호에 있어서 응답자간 선호이질성을 설명하고자 하였다. 6대광역시와 보령시 표본에 대하여 응답자별 이질성의 존재여부를 검토한 결과 두 지역간 뚜렷한 차이가 있음을 알 수 있었다. 즉 보령시의 경우에는 응답자간 선호이질성이 뚜렷하게 나타난데 반하여 6대광역시의 경우 응답자간 선호이질성이 거의 나타나지 않았다. 따라서 6대광역시의 경우에는 모수추정을 위해 CL 모형의 사용이 가능하나 보령시의 경우에는 선호이질성을 반영하기 위해 ML모형이나 LCM에 근거한 모수추정이 요구된다. 선호이질성의 원인을 규명하기 위하여 교차항이 있는 혼합로짓모형과 잠재계층모형을 고려할 수 있다. 교차항이 있는 혼합로짓모형의 경우 관찰되지 않은 개인단위의 이질성을 설명할 수 있는 장점이 있다. 그러나 두모형을 비교한 결과 LCM이 교차항이 있는 ML모형이 제공하지 않는 추가적인 정보를 보여주는 것으로 나타나고 있다. 따라서 본 연구에서의 응답자간 선호이질성은 혼합로짓모형에 의한 개인적인 수준보다 잠재계층모형에 의한 계층단위에서 더 잘 설명될 수 있다고 할 수 있다. Conditional Logit (CL) model is widely used since its model estimation and interpretation of results of the model is relatively easy, on the other hand, it has the limit of preference heterogeneity of respondents being not fully considered. In this study we used the two models, Mixed Logit (ML) Model and Latent Class Model (LCM) to explain preference heterogeneity of respondents for protection for Boryeong Dam wetland. As a result of the examination for heterogeneity in Boryeong city and six metropolitan areas, we found there was significant difference between two regions. While there was explicit preference heterogeneity within respondents in Boryeong city, we found little heterogeneity within respondents in six metropolitan areas. Thus in the case of six metropolitan areas, CL model can be used for parameter estimation while in the case of Boryeong city, WTP estimates are based on parameter estimates from ML model to reflect the heterogeneity within respondents. Additionally, ML model with interaction and 2-class LCM for respondents in Boryeong city were used to explain the sources of the heterogeneity. The ML model with interaction has advantage of explaining individual unobserved heterogeneity. However The comarison between these two models reflects the fact that LCM provided added information that was not conveyed in the ML model with interaction. Thus, Preference heterogeneity within respondents in this study may be better explained by class level through LCM rather than indiviual level through ML model.

ANTECEDENTS OF SUBJECTIVE KNOWLEDGE DEVELOPMENT AND IMPACTS ON FIRST TIME BUYING AND RE-PURCHASING PREFERENCE HETEROGENEITY: EVIDENCE IN THE CHINESE CAR MARKET

Lixian Qian,Didier Soopramanien 글로벌지식마케팅경영학회 2014 Global Marketing Conference Vol.2014 No.7

Subjective knowledge (SK thereafter) is defined as what consumers think they know, or their perceived level of knowledge (Brucks, 1985). SK influences what consumers search for, the effort that they put into searching, and the purchasing decision and choice of brands they finally make (Hadar, Sood, & Fox, 2013; Moorman, Diehl, Brinberg, & Kidwell, 2004). The existing literature on consumer knowledge development mainly argues that SK development is associated with consumers’ product related experience, such as product ownership and experience from usage (Alba & Hutchinson, 2000; Park, Mothersbaugh, & Feick, 1994). This study contributes to the literature as follows. We consider a broad market segmentation where the market consists of product owners and non-owners, The literature suggests that these two groups of consumers should have different levels of perceived knowledge due to their different product-specific experience (Park et al., 1994). Our research specifically contends that it is important to distinguish, within the group of non-owners, between those who intend to buy and those who do not intend to buy the product. Thus, this study examines how previous ownership and intention to buy, which is the goal motivation, influence the SK levels of consumers, which in turn impacts on consumers’ heterogeneous preferences for different attributes within a product category. It is contended that the motivation to learn about a product category of non-car owners who have strong intentions to buy influences what this group learns compared to those non-owners who do not intend to buy. More specifically, the research proposes to uncover how learning influences how different groups of consumers perceive their subjective knowledge and how this relationship in turn shapes their preferences, especially when we compare owners and non-owners who have strong intentions of buying a product. Therefore, we develop following research hypotheses: H1: The SK level is positively associated with the product ownership experience. H2: The SK level is positively associated with purchase intention. H3: The relationship between purchase intention and SK level is moderated by the ownership of the product. H4: Preference heterogeneity for different types of brand attributes of the product between owners and non-owners is moderated by the SK level. The empirical context that illustrates the importance of this research is the Chinese car market where the majority of consumers are first time buyers. Our research shows that the SK level is positively associated with both car ownership and purchase intention. Furthermore, the segmentation analysis finds that purchase intention has a significant effect only on non-car owners’ SK level, while the effect on car owners’ SK level is insignificant. An interesting finding is that SK development, in turn, influences consumer’s preferences for different types of brand attributes of a product. We demonstrate the role of SK in forming consumer preference heterogeneity by comparing preferences for a tangible and a non-tangible brand attribute respectively across different consumer segments. We further show how this preference heterogeneity across two consumer segments (car owners versus non-car owners) depends on their SK level. The results of this study can be generalized beyond the specific context of the empirical study of the Chinese car market when marketers are considering new products launch or launching products where the majority of consumers may be unfamiliar with the product.

성선호 원인에 대한 한국과 자메이카의 비교연구

이성용(Sung Yong Lee) 한국사회학회 2004 韓國社會學 Vol.38 No.3

이 글의 목적은 한국과 자메이카에 존재하는 성선호와 성선호의 원인으로 간주되고 있는 가문계승과 노후보장의 영향을 분석 연구하는 것이다. 한국과 자메이카의 세계출산력 조사를 분석 자료로, 그리고 다단계 위험률 모형을 분석통계 기법으로 사용하였다. 보편 위험률 모형과 비교할 때, 헥크만과 와커(Heckman & Walker, 1992)가 개발한 다단계 위험률 모형은 출발시점에서 비관찰된 개인별 이분산성(unobservable personspecific heterogeneity) 및 가임력에 관련된 변수 그리고 통제변수들을 통제함으로써 후기 출산순위에서의 추정치가 크게 어긋나는 것을 방지할 수 있다. 본 연구의 결과에 따르면 한국은 남아를 선호하고 자메이카는 여아를 선호하는 것으로 나타났다. 기존 자녀의 성구조 변수를 통해 남아선호를 직접적으로 나타낸 한국과는 달리, 자메이카의 여아선호는 자메이카의 모중심 친족제도(matrifocal kinship system)와 연결되어 있다. 성선호의 원인으로 간주되고 있는 가문계승과 노후보장 변수의 통계 분석 결과는 상이한 방향으로 나타났다. 성선호 이론이 가정하는 방향으로 나타난 가문계승변수와는 달리, 노후보장 변수들은 매우 다양한 방향으로 나타났다. 그것은 개인의 가문계승에 대한 개인의 차별적 기대는 가문계승에 대한 사회규범에 의해 지지되지만, 노후보장에 대한 개인의 차별적 기대는 사회적 차원의 노후보장과는 독립적으로 작동하기 때문이다. 가문계승이 노후보장보다 더 중요한 성선호 원인이라는 주장을 하기 전에, 가문계승과 노후보장의 긴밀한 관계가 연구 검토될 필요성을 이 연구는 제시하고 있다. The purpose of this article is to examine the gender preferences of Korea and Jamaica, and to analyze the effect of old age security and family lineage-which are regarded as the causes of gender preference-affecting gender preference. Multi-state-duration-models analyzing data from the World Fertility Survey are used to examine son preference in Korea and daughter preference in Jamaica. Compared with the conventional hazard model, the unobserved person-specific heterogeneity and the explanatory variables are controlled at the initial conditions and the person’s life history relating to fecundity is accounted for in the multi-state-duration-model suggested by Heckman and Walker(1992). The evidences regarding Korea as a son preference country and Jamaica as a daughter preference country are suggested in this paper. However, unlike Korea where son preference is appeared directly through the gender composition of previous children, the daughter preference in Jamaica is represented indirectly through the Jamaican matrifocal family system. The statistical results of family lineage and old age security are not appeared differently. The statistical result of family lineage is consistent of the assumption to which the theories of the cause of gender preference refer. However, the statistical results of old age security are not. Sometimes significant and positively, sometimes significant and negatively, and sometime insignificant. It is so confused. The reasons are following: The different individual expectations about family lineage are supported by the social norms about the family lineage, while those of old age security are independent of the social norm about old age security. In addition, before arguing that family lineage is more important cause of gender preference than old age security, we have to consider the closely connected relationship between family lineage and old age security.

랜덤 패러미터 로짓모델을 이용한 원격대학교 선택요인 분석에 관한 연구

우경봉 한일경상학회 2014 韓日經商論集 Vol.65 No.-

본 연구의 목적은 전국 1,197명으로부터 응답을 받은 설문 데이터를 바탕으로 원격대학교 선택시의 주요 요인을 도출하고, 해당 요인이 원격대학교 선택에 어떠한 영향을 미치는가를 살펴보는 데 있다. 특히, 원격대학교의 브랜드(대학교 이름)가 대학교 선택에 있어 얼마나 강력한 유인이 될 수 있는가에 대해 고찰하기 위해, SP(Stated Preference) 데이터를 수집하였다. 데이터 수집에는 컨조인트 분석을 활용하였으며, 수집된 데이터의 해석에는 개인 간 선호의 다양성(heterogeneity)을 반영할 수 있는 랜덤 패러미터 로짓(Random Parameter Logit)모형을 적용하였다. 연구의 분석결과, 첫째 원격대학교 선택의 주요 요인들로는 등록금, 교육의 질, 졸업난이도 등이 도출되었다. 둘째, 원격대학교 선택시 등록금, 오프라인 강의 횟수, 학습난이도 등에 대해서는 해당 변인의 증가를 원하지 않으며, 시험은 공공장소에서의 감독관 입회 시험보다 재택시험을 선호하는 것으로 나타났다. 셋째, 연령이 증가할수록 원격대학교에 대한 선택확률은 낮아졌으며, 남성보다 여성이 원격대학교에 높은 관심을 가지고 있었다. 또한 소득이 증가할수록 원격대학교에서의 교육에 높은 관심을 보였다. 넷째, 원격대학교 중 기존의 사회적 인지도가 가장 높은 곳은 A대학교이지만, 원격대학교의 주요 속성의 결합 정도에 따라서는 대학교 브랜드의 유인력이 약화될 수 있다. 이러한 결과는 원격대학교 관련 정책 및 경쟁력 제고의 방향성 논의에 시사하는 바가 클 것으로 생각된다. The purpose of this study is to analyze the factors of distance university choice and to compare the brand powers of distance universities of Republic of Korea. To carry out the research conjoint survey method was employed to collect stated preference data for 1197 respondents who live in Republic of Korea. In analyzing the data, Random Parameter logit model was used not only to estimate the parameters’ means and standard deviations but also to estimate the kernel density for the preference of respondents’ heterogeneity. Analysis results show that fist, main factors of distance university choice are tuition, the quality of education at distance university, the level of difficulty for graduation etc.. Second, they don’t want to increase in tuition, off-line classes, the level of difficulty for study. And they prefer to take midterm and final term by their own PCs. Third, women have tendency to be more interested in distance university than men. Forth, according to combinations of the distance university choice factors, the strength of brand powers of distance universities might be changed.

바이오에탄올 보급에 대한 사회적 수용성 분석: 바이오에탄올 원료 속성을 중심으로

이드미트리 ( Dmitriy D. Li ),배정환 ( Jeong Hwan Bae ) 한국환경경제학회·한국자원경제학회(구 한국환경경제학회) 2021 자원·환경경제연구 Vol.30 No.1

본 연구는 선택실험법을 이용하여 바이오에탄올 원료유형, 바이오에탄올 혼합율, 바이오에탄올 비용, 정부지원 정책과 같은 속성들이 바이오에탄올 보급정책에 대한 사회적 수용성에 영향을 미치는지를 분석하였다. 바이오에탄올 속성 계수를 추정하기 위해 조건부로짓모형, 혼합로짓모형, 개별계수추정모형을 적용하였다. 추정 결과에 따르면, 소비자들은 국산원료와 비식량원료를 사용한 바이오에탄올을 선호하고 지원정책 가운데는 면세정책을 선호하는 것으로 나타났다. 혼합로짓모형과 개별계수추정모형에 의하면 모든 속성들이 관찰불가능한 이질성을 갖고 있는 것으로 나타났다. 또한 속성별 지불용의액을 추정한 결과, 유전자조작기반 바이오 에탄올임을 사전에 인지한 응답자일수록 그렇지 못한 응답자보다 바이오에탄올에 대한 지불용의액이 더 낮게 나타났다. 추정결과를 종합하면, 우리나라에서 바이오에탄올을 보급하기 위해서는 국산원료 및 비식량원료에 기반한 바이오에탄올을 중점적으로 보급해야 하고, 특히 유전자 조작 기반 바이오에탄올에 대한 수용성이 낮게 나타나므로 보급시 이를 충분히 고려해야 할 것이다. This study uses a choice experiment approach to examine whether different types of feedstocks as well as other attributes such as the cost of bioethanol, bioethanol blending ratio, and government support policies affect consumers’ biofuel preferences. We apply a standard conditional logit model, a mixed logit model (MLM), and individual coefficient estimation model (ICM) to estimate the parameters of the investigated attributes. The results show that people prefer domestic and non-food feedstock, along with tax exemption as a support policy. All the attributes show unobservable preference heterogeneity in the MLM and ICM. In particular, willingness to pay for attributes are higher in the genetically modified (GM) feedstock-unknown group than in the known one. We show the importance of using domestic and non-food feedstocks and managing GM feedstocks carefully to avoid consumer resistance when producing bioethanol in South Korea.

생산지역이 사과와 사과주스 구매에 미치는 영향

김현정,김선웅 한국농업경제학회 2022 農業經濟硏究 Vol.63 No.2

The objective of this study is to elicit the region-of-origin effect on apple and its juice. The questionnaire was structured based on choice experiment (CE). Respondents were asked to choose among four alternatives, each described by varying the prices, the cultivation methods, the production areas, and the store types. Respondents were randomly assigned to treatment and control groups with/without offering social benefits from using locally grown food. The conditional logit and latent class model were applied to estimate the consumers' willingness-to-pay (WTP) for the attributes with considering preference heterogeneity. The results are as follows: The WTP for locally produced apple is higher than that for domestic one only in the treatment group. It means that providing the information on the social benefit of local foods can increase the purchase intention. Consumers can be divided into two groups: the first considers the cultivation environment such as the cultivation methods, the production areas more important than the price, and the second one does the price than others when purchasing agricultural products.

개인의 노력 및 선호를 반영한 불평등 측정:등가소득 분석

박강우,서영선 한신대학교 민주사회정책연구원 2016 민주사회와 정책연구 Vol.0 No.31

Employing Korean Labor and Income Panel Study(KLIPS) data, we evaluate the equivalent income of individual wage-earners, which indicates the welfare level accounting for the difference in individual effort and leisure-labor preference. Comparing the distribution between the observed income and the equivalent income, we find that the distribution of the equivalent income, reflecting the disutility from labor effort measured by hours worked, is more unequal than that of the observed income over the whole sample period. The reason is two-fold: high-income earners, on average, work less hours than middle-or-low-income earners, and there are huge differences in hours worked among the workers earning the same level of income. In addition to hours worked, when educational achievement is accounted for as another factor representing individual effort, there is no qualitative change in our main results. While the degree of inequality in the observed income is stable across the sample period, that of inequality in the equivalent income is strikingly mitigated after the Global Financial Crisis. When we estimate the determining factors of inequality, the degree of inequality by father’s or householder’s educational achievement as well as that of one’s own is markedly higher in the equivalent income than in the observed income, while the degree of inequality by sex is lower in the equivalent income than in the observed income. This implies that the well-bred or well-educated workers tend to earn higher income and simultaneously work less. 본고에서는 우리나라 노동패널 데이터를 이용하여 근로시간으로 대표되는 개인의 노력수준과 여가-노동에 대한 선호 차이를 반영한 후생수준을 나타내는 등가소득을 추정하고, 등가소득과 실제 관측소득의 분포 및 불평등도를 비교하였다. 분석 결과, 전체 표본기간에 걸쳐 근로의 비효용을 반영한 후생수준을 나타내는 등가소득의 불평등도가 관측소득에 비해 높은 것으로 나타났다. 이는 우리나라의 경우 고소득층의 근로시간이 중저소득층의 근로시간에 비해 평균적으로 짧을 뿐만 아니라, 같은 소득을 버는 개인 간에도 근로시간의 격차가 매우 크기 때문인 것으로 판단된다. 또한 개인의 노력을 대표하는 변수로서 근로시간과 함께 교육수준을 고려한 경우에도 불평등도의 절대수준은 소폭 감소하나 여전히 등가소득의 불평등도가 관측소득에 비해 높았다. 한편 시간에 따라 불평등도가 대체로 일정한 관측소득과 달리 등가소득에 따른 불평등도는 금융위기 이후 최근으로 오면서 뚜렷하게 줄어드는 모습을 보였다. 아울러 연도별 회귀분석을 통해 등가소득과 관측소득의 결정요인을 추정·비교한 결과, 성별 소득격차의 경우 등가소득이 관측소득에 비해 불평등도가 낮은 것으로 나타났으나, 본인이나 아버지 또는 가구주의 학력에 따른 소득격차의 경우 등가소득이 관측소득에 비해 불평등도가 전반적으로 높게 나타났다. 이는 본인의 학력과 가정배경 등 환경요인에 따라 고소득을 벌어들이는 직업을 가진 사람은 대체로 근로시간도 짧은 경향이 있음을 의미한다.

개인의 노력 및 선호를 반영한 불평등 측정: 등가소득 분석

박강우 ( Park Kangwoo ),서영선 ( Seo Youngseon ) 민주사회정책연구원 2017 민주사회와 정책연구 Vol.31 No.-

본고에서는 우리나라 노동패널 데이터를 이용하여 근로시간으로 대표되는 개인의 노력수준과 여가-노동에 대한 선호 차이를 반영한 후생수준을 나타내는 등가소득을 추정하고, 등가소득과 실제 관측소득의 분포 및 불평등도를 비교하였다. 분석 결과, 전체 표본기간에 걸쳐 근로의 비효용을 반영한 후생수준을 나타내는 등가소득의 불평등도가 관측소득에 비해 높은 것으로 나타났다. 이는 우리나라의 경우 고소득층의 근로시간이 중저소득층의 근로시간에 비해 평균적으로 짧을 뿐만 아니라, 같은 소득을 버는 개인 간에도 근로시간의 격차가 매우 크기 때문인 것으로 판단된다. 또한 개인의 노력을 대표하는 변수로서 근로시간과 함께 교육수준을 고려한 경우에도 불평등도의 절대수준은 소폭 감소하나 여전히 등가소득의 불평등도가 관측소득에 비해 높았다. 한편 시간에 따라 불평등도가 대체로 일정한 관측소득과 달리 등가소득에 따른 불평등도는 금융위기 이후 최근으로 오면서 뚜렷하게 줄어드는 모습을 보였다. 아울러 연도별 회귀분석을 통해 등가소득과 관측소득의 결정요인을 추정·비교한 결과, 성별 소득격차의 경우 등가소득이 관측소득에 비해 불평등도가 낮은 것으로 나타났으나, 본인이나 아버지 또는 가구주의 학력에 따른 소득격차의 경우 등가소득이 관측소득에 비해 불평등도가 전반적으로 높게 나타났다. 이는 본인의 학력과 가정배경 등 환경요인에 따라 고소득을 벌어들이는 직업을 가진 사람은 대체로 근로시간도 짧은 경향이 있음을 의미한다. Employing Korean Labor and Income Panel Study(KLIPS) data, we evaluate the equivalent income of individual wage-earners, which indicates the welfare level accounting for the difference in individual effort and leisure-labor preference. Comparing the distribution between the observed income and the equivalent income, we find that the distribution of the equivalent income, reflecting the disutility from labor effort measured by hours worked, is more unequal than that of the observed income over the whole sample period. The reason is two-fold: high-income earners, on average, work less hours than middle-or-low-income earners, and there are huge differences in hours worked among the workers earning the same level of income. In addition to hours worked, when educational achievement is accounted for as another factor representing individual effort, there is no qualitative change in our main results. While the degree of inequality in the observed income is stable across the sample period, that of inequality in the equivalent income is strikingly mitigated after the Global Financial Crisis. When we estimate the determining factors of inequality, the degree of inequality by father`s or householder`s educational achievement as well as that of one`s own is markedly higher in the equivalent income than in the observed income, while the degree of inequality by sex is lower in the equivalent income than in the observed income. This implies that the well-bred or well-educated workers tend to earn higher income and simultaneously work less.

Exploring the Role of Preference Heterogeneity and Causal Attribution in Online Ratings Dynamics

주우진,노민정 한국마케팅학회 2014 ASIA MARKETING JOURNAL Vol.15 No.4

This study investigates when and how disagreements in online customer ratings prompt more favorable product evaluations. Among the three metrics of volume, valence, and variance that feature in the research on online customer ratings, volume and valence have exhibited consistently positive patterns in their effects on product sales or evaluations (e.g., Dellarocas, Zhang, and Awad 2007; Liu 2006). Ratings variance, or the degree of disagreement among reviewers, however, has shown rather mixed results, with some studies reporting positive effects on product sales (e.g., Clement, Proppe, and Rott 2007) while others finding negative effects on product evaluations (e.g., Zhu and Zhang 2010). This study aims to resolve these contradictory findings by introducing preference heterogeneity as a possible moderator and causal attribution as a mediator to account for the moderating effect. The main proposition of this study is that when preference heterogeneity is perceived as high, a disagreement in ratings is attributed more to reviewers’ different preferences than to unreliable product quality, which in turn prompts better quality evaluations of a product. Because disagreements mostly result from differences in reviewers’ tastes or the low reliability of a product’s quality (Mizerski 1982; Sen and Lerman 2007), a greater level of attribution to reviewer tastes can mitigate the negative effect of disagreement on product evaluations. Specifically, if consumers infer that reviewers’ heterogeneous preferences result in subjectively different experiences and thereby highly diverse ratings, they would not disregard the overall quality of a product. However, if consumers infer that reviewers’ preferences are quite homogeneous and thus the low reliability of the product quality contributes to such disagreements, they would discount the overall product quality. Therefore, consumers would respond more favorably to disagreements in ratings when preference heterogeneity is perceived as high rather than low. This study furthermore extends this prediction to the various levels of average ratings. The heuristic- systematic processing model so far indicates that the engagement in effortful systematic processing occurs only when sufficient motivation is present (Hann et al. 2007; Maheswaran and Chaiken 1991; Martin and Davies 1998). One of the key factors affecting this motivation is the aspiration level of the decision maker. Only under conditions that meet or exceed his aspiration level does he tend to engage in systematic processing (Patzelt and Shepherd 2008; Stephanous and Sage 1987). Therefore, systematic causal attribution processing regarding ratings variance is likely more activated when the average rating is high enough to meet the aspiration level than when it is too low to meet it. Considering that the interaction between ratings variance and preference heterogeneity occurs through the mediation of causal attribution, this greater activation of causal attribution in high versus low average ratings would lead to more pronounced interaction between ratings variance and preference heterogeneity in high versus low average ratings. Overall, this study proposes that the interaction between ratings variance and preference heterogeneity is more pronounced when the average rating is high as compared to when it is low. Two laboratory studies lend support to these predictions. Study 1 reveals that participants exposed to a high-preference heterogeneity book title (i.e., a novel) attributed disagreement in ratings more to reviewers’ tastes, and thereby more favorably evaluated books with such ratings, compared to those exposed to a low-preference heterogeneity title (i.e., an English listening practice book). Study 2 then extended these findings to the various levels of average ratings and found that this greater preference for disagreement options under high preference heterogeneity is more pronounced when the average rating is hig...

선택실험법을 이용한 중장기 전원별 소비자 선호 분석

정해영,배정환 한국환경경제학회 2018 자원·환경경제연구 Vol.27 No.4

Recently, extreme weather due to climate change has become more frequent, and increase of fine dust has worsen air quality in Korea. Therefore, not only negative perception on coal-fired power generation is dominant, but also the social acceptance of nuclear power generation declines. This study aims at deriving consumer preferences on the mid and long term power mix with various energy sources. Willingness to pay for each generation source was estimated and the preference heterogeneity of consumers was examined by using mixed logit and latent class models. Mixed logit estimation results show that the preference heterogeneity of consumers is especially large for the nuclear power relative to renewable or coal energy. According to the estimation results from the latent class model, group 1 prefers renewable energy while group 2 prefers coal energy. Group 3 shows lexicographic preference which means restricted rationality. As for the policy implication, it is necessary to understand the preference heterogeneity of consumer groups in planning the mid to long term power mix. 최근 들어 기후변화로 인한 이상기후가 빈발해지고, 미세먼지 문제가 심각해짐에 따라 석탄화력발전에 대한 부정적인 인식이 커지고 있고, 경주 지진 이후 원자력 발전에 대한 사회적 수용력도 감소하고 있다. 본 연구는 선택실험법을 이용하여 중장기 전원별 일반 소비자들의 지불용의액을 도출함으로써 각 발전원에 대한 경제적 가치를 추정하고, 소비자들의 선호 이질성을 검토하였다. 적용한 방법론은 혼합로짓과 잠재계층모형이며, 혼합로짓 추정결과 특히 원자력에서 선호이질성이 크게 나타나는 것을 확인하였다. 잠재계층모형 추정결과 집단1은 신재생에너지를 선호하며, 집단2는 기존의 화석에너지를 선호하였다. 집단3은 사전편찬식 선호(제한적 합리성)을 보이는 집단으로 추정되었다. 사전편찬식 선호를 보이는 집단을 고려하지 않을 경우 지불용의액 추정의 신뢰성이 낮아지고, 속성의 유의성이 낮아지는 결과가 나타나는 것을 확인하였다. 또한 집단별로 에너지원에 대한 선호 이질성이 크게 나타나, 향후 전원구성에 있어 전력 소비자의 선호 이질성을 감안한 정책이 필요할 것이다.