• The Korean Journal of Applied Statistics
• /
• v.32 no.2
• /
• pp.173-185
• /
• 2019

Causal questions are prevalent in scientific research, for example, how effective a treatment was for preventing an infectious disease, how much a policy increased utility, or which advertisement would give the highest click rate for a given customer. Causal inference theory in statistics interprets those questions as inferring the effect of a given intervention (treatment or policy) in the data generating process. Causal inference has been used in medicine, public health, and economics; in addition, it has received recent attention as a tool for data-driven decision making processes. Many recent datasets are observational, rather than experimental, which makes the causal inference theory more complex. This review introduces key concepts and recent trends of statistical causal inference in observational studies. We first introduce the Neyman-Rubin's potential outcome framework to formularize from causal questions to average treatment effects as well as discuss popular methods to estimate treatment effects such as propensity score approaches and regression approaches. For recent trends, we briefly discuss (1) conditional (heterogeneous) treatment effects and machine learning-based approaches, (2) curse of dimensionality on the estimation of treatment effect and its remedies, and (3) Pearl's structural causal model to deal with more complex causal relationships and its connection to the Neyman-Rubin's potential outcome model.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.615-632
• /
• 2017

Unlike randomized trial, statistical strategies for inferring the unbiased causal relationship are required in the observational studies. The matching with the propensity score is one of the most popular methods to control the confounders in order to evaluate the effect of the treatment on the outcome variable. Recently, new methods for the causal inference in latent class analysis (LCA) have been proposed to estimate the average causal effect (ACE) of the treatment on the latent discrete variable. They have focused on the application study for the real dataset to estimate the ACE in LCA. In practice, however, the true values of the ACE are not known, and it is difficult to evaluate the performance of the estimated the ACE. In this study, we propose a method to generate a synthetic data using the propensity score in the framework of LCA, where treatment and outcome variables are latent. We then propose a new method for estimating the ACE in LCA and evaluate its performance via simulation studies. Furthermore we present an empirical analysis based on data form the 'National Longitudinal Study of Adolescents Health,' where puberty as a latent treatment and substance use as a latent outcome variable.

• The Korean Journal of Applied Statistics
• /
• v.36 no.1
• /
• pp.63-83
• /
• 2023

Many existing studies have found significant correlations between green spaces, including urban parks, and children's happiness. Furthermore, it was implied that the area/proximity of the urban park would be effective in enhancing infancy happiness. However, inferring causal effects from observed data requires appropriate adjustment of confounding variables, and from this perspective, the causal relationship between the area of urban parks and children's happiness has not been well understood. The causal effect of urban parks on children's happiness was estimated in this study using data from the panel study on Korean children. As methods for adjusting confounding variables, regression adjustment using a regression method, weighting method, and matching method were used, and key concepts of each method were described before the analysis results. Confounders were chosen for the analysis using a directed acyclic graph. In contrast to previous research, the analysis found no significant causal relationship between the size of the city park and children's happiness.

• Environmental and Resource Economics Review
• /
• v.5 no.2
• /
• pp.279-290
• /
• 1996

정부의 환경규제로 인하여 한국 제조업의 기업들이 공해방지시설을 의무적으로 설치해야할 경우, 이는 생산비증가와 신규투자의 위축 및 생산요소간 조합비율에 영향을 미침으로써 직 간접적으로 생산성의 감소를 초래한다. 본 연구에서는 인과추정법(因果推定法)으로 측정한 그레이 (Gray, 1987)의 모형을 통하여 환경규제가 한국 제조업의 생산성에 미치는 효과를 측정하였다. 측정결과, 규제로 인하여 연평균 0.58퍼센트 포인트만큼의 생산성감소효과가 있었다. 여기에는 생산성증가율 계산상의 '측정효과(測定效果)'만 존재하고 생산과정의 제약, 신규투자의 위축, 다른 생산요소의 생산성감소로 인하여 총생산성이 감소되는 '실질효과(實質效果)'는 발생하지 않았다.

• Journal of Labour Economics
• /
• v.42 no.4
• /
• pp.59-88
• /
• 2019

This study estimates causal effects of the governor vacancy on the regional economy, exploiting the case of exogenous governor vacancy in Korea. We find that the governor vacancy has a negative impact on the regional economy by lowering the employment rate and reducing the amount of credit card expenditures. Negative effects are more pronounced among vulnerable groups of the labor market such as women and aged 20~29 and 50~59. In addition, negative effects vary by characteristics of the governor. Negative effects of the governor vacancy on the regional economy show empirical evidence suggesting that leaders do matter.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.124-124
• /
• 2017

수자원 및 수도 시설과 같은 사회간접자본은 경제개발 초기 단계에 있어 매우 중요한 역할을 수행하여 왔다. 우리나라의 경제성장 과정에서 수자원 개발은 "한강의 기적"으로 불리며 한국 경제발전의 중요한 원동력 가운데 하나로 알려져 있다. 본 연구에서는 수자원개발의 경제적 효과를 정량적으로 분석하기 위하여 거시경제모형을 구축하고 실증분석하고자 한다. 이를 위해 1977-2014년 동안의 수자원 부문에 대한 자본 스톡을 추정하고 이를 바탕으로 경제성장과의 인과관계를 검정한다. 추정결과, 수자원 투자는 경제성장(GDP)으로의 단방향의 인과성이 존재함을 확인(1%유의 수준)하였다. 외생적 충격으로 수자원 투자가 감소하는 경우 국내 소득(GDP)에 부정적 영향을 미칠 수 있음을 시사하고 있다. 우리나라의 성공적인 수자원 개발과 경제 발전 경험은 많은 개도국에게 시사점을 제공할 것이다.

• Journal of Labour Economics
• /
• v.39 no.2
• /
• pp.1-34
• /
• 2016

We estimate the effect of an increase in the minimum wage on employment. In Korea, there is no exogenous variation in the minimum wage across regions or industries. One single minimum wage is applied to every worker in the whole country. In this paper, we exploit arguably exogenous variation in the proportion of workers affected by the minimum wage across worker groups defined by age, sex, education, tenure and establishment size. Using the data from the Survey on Labor Conditions by Type of Employment (SLCTE) from 2006 to 2014, we find that a 1% increase in the minimum wage decreases the full-time equivalent employment by about 0.14%. The effect is heterogeneous across workers; we find the effect is more adverse for female workers, low-educated, younger and older workers, workers with a shorter tenure, and workers in small- and medium-sized establishments.

• The Korean Journal of Applied Statistics
• /
• v.30 no.6
• /
• pp.943-955
• /
• 2017

We investigate and estimate the causal effect of the survey methods in telephone surveys for the 19th presidential election. For this causal study, we draw a causal graph that represents the causal relationship between variables. Then we decide which variables should be included in the model and which variables should not be. We explain why the research agency is a should-be variable and the response rate is a shouldnot-be variable. The effect of ARS can not be estimated due to data limitations. We have found that there is no significant difference in the effect of the proportion of cell phone survey if it is less than about 90 percent. But the support rate for Moon Jae-in gets higher if the survey is performed only by cell phones.

• The Korean Journal of Applied Statistics
• /
• v.28 no.5
• /
• pp.871-883
• /
• 2015

Vector autoregressive (VAR) models in high dimension suffer from noisy estimates, unstable predictions and hard interpretation. Consequently, the sparse vector autoregressive (sVAR) model, which forces many small coefficients in VAR to exactly zero, has been suggested and proven effective for the modeling of high dimensional time series data. This paper studies coupling measures to select non-zero coefficients in sVAR. The basic idea based on the simulation study reveals that removing the effect of other variables greatly improves the performance of coupling measures. sVAR model coefficients are asymmetric; therefore, asymmetric coupling measures such as Granger causality improve computational costs. We propose two asymmetric coupling measures, filtered-cross-correlation and filtered-Granger-causality, based on the filtered residuals series. Our proposed coupling measures are proven adequate for heavy-tailed and high order sVAR models in the simulation study.

• Management & Information Systems Review
• /
• v.39 no.2
• /
• pp.17-37
• /
• 2020

The purpose of this study is to investigate the global spillover effects through the existence of linear and nonlinear causal relationships between the US, European and BRIC financial markets after the period from the introduction of the Euro, the financial crisis and the subsequent EU debt crisis in 2007~2010. Although the global spillover effects of the financial crisis are well described, the nature of the volatility effects and the spread mechanisms between the US, Europe and BRIC stock markets have not been systematically examined. A stepwise filtering methodology was introduced to investigate the dynamic linear and nonlinear causality, which included a vector autoregressive regression model and a multivariate GARCH model. The sample in this paper includes the post-Euro period, and also includes the financial crisis and the Eurozone financial and sovereign crisis. The empirical results can have many implications for the efficiency of the BRIC stock market. These results not only affect the predictability of this market, but can also be useful in future research to quantify the process of financial integration in the market. The interdependence between the United States, Europe and the BRIC can reveal significant implications for financial market regulation, hedging and trading strategies. And the findings show that the BRIC has been integrated internationally since the sub-prime and financial crisis erupted in the United States, and the spillover effects have become more specific and remarkable. Furthermore, there is no consistent evidence supporting the decoupling phenomenon. Some nonlinear causality persists even after filtering during the investigation period. Although the tail distribution dependence and higher moments may be significant factors for the remaining interdependencies, this can be largely explained by the simple volatility spillover effects in nonlinear causality.