Earticle

환경쿠즈네츠 곡선(EKC)은 환경오염원과 소득간의 관계를 역U자형(inverted U shape) 곡선으로 설명하는 가설이다. 본 연구에서는 환경쿠즈네츠 곡선(EKC)의 가능성을 추정하기 위해 기존연구와 달리 국가의 1차 에너지소비량을 오염원의 대리변수로 선정하여 65개 국가를 대상으로 2단계 추정전략을 적용하였다. 1단계에서는 65개 국가의 1971년∼2010년의 1차 에너지소비량을 오염원의 대리변수로 사용하여 개별국가를 대상으로 이동회귀모형(rolling regression)을 추정하여 국가 및 기간별 EKC의 성립여부에 따라 이항변수를 생성하였다. 2단계에서는 1단계에서 생성한 이항변수를 종속변수로 설정 후 EKC를 가능하게 하는 주요 설명변수를 결정요인으로 선정하여 패널이항모형(panel binary model)을 추정하였다. 추정결과, 도시인구비율증가, 에너지효율성개선, 순에너지수입증가, 정치적 평등도 개선은 EKC 가설이 성립할 확률을 증가시켰으나 GDP 대비 상품 및 서비스 수입비율 증가와 2차산업비중 증가는 EKC 가설이 성립할 확률을 감소시켰다.

The Environmental Kuznets Curve(EKC) hypothesis represents the inverted U shape relationship between economic growth and pollution. Unlike previous literature, we use primary energy consumption as the broad proxy of pollution. At the first stage, we examines the inverted U shape for each country by rolling regression. Following the first stage, we estimates the probability of EKC with major regressors using panel binary model. It turns out that urban population ratio, energy efficiency, net energy import, and political equality have a positive effect on EKC, while import to GDP ratio and secondary industry share have a negative effect on EKC.

The fixed effects panel probit model faces "incidental parameters problem" because it has a property that the number of parameters to be estimated will increase with sample size. The maximum likelihood estimation fails to give a consistent estimator of slope parameter. Unlike the panel regression model, it is not feasible to find an orthogonal reparameterization of fixed effects to get a consistent estimator. In this note, a hierarchical Bayesian model is proposed. The model is essentially equivalent to the frequentist's random effects model, but the individual specific effects are estimable with the help of Gibbs sampling. The Bayesian estimator is shown to reduce reduced the small sample bias. The maximum likelihood estimator in the random effects model is also efficient, which contradicts Green (2004)'s conclusion.

Panel data sets have recently been developed in various areas, and many recent studies have analyzed panel, or longitudinal data sets. Often a dichotomous dependent variable occur in survival analysis, biomedical and epidemiological studies that is analyzed by a generalized linear mixed effects model (GLMM). The most common estimation method for the binary panel data may be the maximum likelihood (ML). Many statistical packages provide ML estimates; however, the estimates are computed from numerically approximated likelihood function. For instance, R packages, pglm (Croissant, 2017) approximate the likelihood function by the Gauss-Hermite quadratures, while Rchoice (Sarrias, Journal of Statistical Software, 74, 1-31, 2016) use a Monte Carlo integration method for the approximation. As a result, it can be observed that different packages give different results because of different numerical computation methods. In this note, we discuss the pros and cons of numerical methods compared with the exact computation method.

Government has made policy efforts to ease intergenerational inequality by implementing financial and tax policies intended for the first house purchasers in life or the houseless and by enforcing the merit system for newly-wedded couples. But in reality a considerable number of households in a dead zone of high housing price and welfare policy have not benefited from government policy. Therefore, this study attempted to analyze the branch families of newly-wedded couples with poor assets with the use of the ‘panel probit model’ in order to investigate the effect of their socioeconomic characteristics on the determination of housing occupation types. As a result, it was found that the income and asset variables of their parents did not have a significant effect on their children' determination of housing occupation types but such variables as the educational level and occupation(employer) had a significant effect on it. And analysis showed that the characteristic of parents’ household had an insignificant effect on their child’s determination of housing occupation type. This study, unlike previous studies, showed that the branch families of newly-wedded couples with lower educational levels had an increasing rate of homeownership. This is thought to be due to the background of the times that they fall into an age range of economically stable households due to long-standing income regardless of their educational level. In particular, number of children and type of occupation of original household turn out to be important factors in housing tenure choice for married-off households, which implies the possibility of intergenerational wealth transfer.

국무총리는 국정을 통합적으로 관리하기 위하여 정부업무평가기본법에 근거하여 중앙행정기관의정책 등을 평가하고 있고, 중앙행정기관은 정부업무평가기본법에 근거하여 성과관리 등을 실시하면서자체평가를 실시하고 있다. 동 제도가 시행된 지 18년이 지났으나, 자체평가와 정부업무평가결과와연계성을 확인하는 실증 연구는 찾아보기 어렵다. 본 연구는 2018년부터 2022년까지 국무조정실이주관하는 정부업무평가 결과와 각 부처가 시행하는 자체평가 결과에 대해 국무조정실이 메타평가를하는 성과관리 운영 실태 점검 결과 자료와 행정안전부가 실시하는 행정 관리 역량 점검 결과가 정부업무평가의 특정평가(이하 특정평가)에 미치는 영향에 대하여 패널 순서형 프로빗 모형(Panel Ordered Probit model)을 이용하여 분석하였다. 연구자는 양 제도의 연계성을 자체평가가 충실한 기관이 정부업무평가에도 우수한 성과를 이루고있는지에 대해 초점을 맞추어 분석하였다. 동 자료를 종단적으로 분석한 결과 자체평가 우수기관이정부업무평가 결과 우수기관으로 유지되고 있는 사실이 발견되었다. 아울러, 패널 회귀분석(Panel regression)과 상관분석 등을 통하여 정부업무평가 특정평가의 주요구성인 주요정책평가 결과가 정부업무평가 특정평가 종합등급에 가장 큰 영향을 주고 있는 사실을확인하였다. 동 연구는 특정평가와 자체평가를 실시한 중앙행정기관의 업무에 대한 평가결과를 활용하여 평가 부문별 상호 연계성을 도출했다는 점에서 의의가 있다고 할 것이다.

The Prime Minister evaluates the policies of central administrative agencies based on the Framework Act on Government Performance Evaluation to manage national affairs in an integrated manner, and the central administrative agencies conduct self-evaluation by conducting performance management, etc. based on the Framework Act on Government Performance Evaluation. Although 18 years have passed since the system was implemented, it is difficult to find empirical studies on connectivity between the government performance evaluation and the self-evaluation results. In this study, the impact of the management competency inspection results on the specific evaluation of government performance evaluation (hereinafter referred to as specific evaluation) was analyzed using the Panel Ordered Probit model. The analysis used the results of the performance management operation status inspection, meta-evaluation of the results of the government performance evaluation, conducted by the Office for Government Policy Coordination and the result of the performance management operation status inspection conducted by the Ministry of Public Administration and Security among the results of the self-evaluation conducted by each ministry, from 2018 to 2022,. The researcher analyzed the connectivity between the two systems, focusing on whether the government agencies that had showed good self-evaluation results were also achieving excellent performance in government performance evaluation. As a result of longitudinal analysis of this data, it was discovered that excellent self-evaluated agencies were also had excellent result on government performance evaluation. In addition, it was confirmed that main policy evaluation result, which was the main component of the specific evaluation, had the greatest impact on the comprehensive grade of the specific evaluation through panel regression and correlation analysis. This study is significant in that it proves interconnection between evaluation sectors by using specific evaluation and evaluation results of the work of central administrative agencies that conducted self-evaluations.

경시적 자료는 같은 개체에서 시간에 따라 반복 측정된 자료이다. 따라서 반복 측정된 자료는 상관관계가 존재하며 이것을 설명하면서 공변량의 반응변수의 효과를 추정해야 한다. 경시적 순서형 자료분석에서는 잠재변수의 조건부 누적확률을 로짓 연결함수 또는 프로빗 연결함수를 이용한 선형혼합 모형을 이용하여 공변량의 효과를 추정한다. 본 논문에서는 경시적 순서형 자료분석을 위한 두 가지 형태의 연결함수를 가지는 일반화선형혼합모형 및 주변화모형을 고찰한다. 그리고 최근에 제안된 프로빗 연결함수를 가지는 베이지안 누적 프로빗 선형혼합모형을 이용하여 경시적 순서형자료인 사업체 패널조사자료를 분석한다. 이 모형은 잠재변수의 조건부 상관계수 행렬의 모형화에 초구분해를 고려하여 고차원이며 양정치성을 만족하는 상관계수를 추정하는 방법이다. 사업체 패널 조사자료는 반응변수로 순서형 자료인 사업체의 교육훈련참여율을 고려하였고, 상관계수 행렬은 자기상관구조를 가정한 여러 모형을 비교하고 가장 적합한 모형을 제시한다. 그리고 그 모형을 이용하여 연도별 효과와 성과배분제도 실시여부, 1인당 연평균 교육시간, 노동조합여부가 유의미한 것을 찾았다.

Longitudinal data are measured repeatedly over time from the same subject. Therefore, the repeated outcomes have correlations, and it is necessary to estimate the covariate effect on the response variable while explaining the correlations. In longitudinal ordinal data analysis, the covariate effect is estimated using generalized linear mixed models using a logit link function or a probit link function. In this paper, we review the generalized linear mixed models and marginalized models with the two types of link functions for longitudinal ordinal data analysis. Specifically, a Bayesian cumulative probit linear mixed model with the probit link function is used to analyze Korean workplace panel survey (WPS) data, which is longitudinal ordinal data. In the model, the correlation matrix is high-dimensional and positive definite, and it is estimated using the hypersphere decomposition. In the WPS data, corporate training participation rate is considered as a response variable. Assuming different correlation structures, several models are compared. For the most suitable model, some explanatory variables, the annual effect, profit sharing schemes status, average annual training hours per person, and labor union status, have effects on corporate training participation rate.