• Survey Research
• /
• v.10 no.3
• /
• pp.21-36
• /
• 2009

This study suggests the applicability of web surveys regarding elections in order to contact a great number of young people. The propensity weighting model was estimated using the demographic variables and the covariate variables collected during the 2007 presidential election surveys. In order to adjust the internet survey to the telephone survey, we used the propensity score method. Propensity score weighting made the internet survey results closer to the telephone survey results. This shows that an internet survey with propensity weighting model is a potential alternative survey method in the prediction of elections.

• Survey Research
• /
• v.10 no.1
• /
• pp.169-186
• /
• 2009

The propensity score method is used to minimize the bias level in social survey, which comes from nonresponse. The theoretical concept and the background of the propensity score method is discussed first. The propensity score method was first applied in the epidemiology observational study. I have summarized the process of the three propensity score methods that were used to reduce estimation bias in this study. Matching by propensity score is applied to the relatively large control group. Subclassification has the advantage of using whole control group data and regression adjustment is applied to multiple covariates as well as propensity score of each unit is computable and usable. Lastly, the application procedures of propensity score method to reduce the nonresponse bias is suggested and its applicability to real situation is reviewed with the existing data.

• Survey Research
• /
• v.11 no.2
• /
• pp.1-28
• /
• 2010

This paper reports the results of the 2009 Internet volunteer panel version of the social survey conducted by Statistics Korea (Korea National Statistical Office). Authors identify socio-psychological characteristics of Internet survey volunteers and present quantitative evaluation of the propensity adjustment weighting method intended to remove Internet sample bias. The nine criteria used for propensity adjustment were regions, urban/rural, gender, age, education, consumer satisfaction, views on income distribution, newspaper access and Internet news access. Propensity adjustment weighting based on the logit model and rim weights were applied to the online survey of 2,903 respondents using the face-to-face area sample data of 37,049 respondents as reference. A total of 106 items were used for evaluating the propensity adjustment weighting methods. The results showed that in 80% of survey items the propensity adjustment weighting yielded better estimates compared to simple demographic weighting. This suggests that Internet surveys by volunteer panels are useful for conducting the general social study in Korea. The reference survey data for this study contains several items on social-psychological behaviors and attitudes, is large in size and obtained by probability sampling. Thus it may be utilized in propensity adjustment of other Internet surveys.

• Proceedings of the Korea Water Resources Association Conference
• /
• 1990.07a
• /
• pp.29-39
• /
• 1990

측방유입수가 고려되는 3변수 Muskingum하도추적모형을 낙동강수계중 왜관에서 적포교구간의 12개 홍수사상에 대하여 적용하였고, 기존방법인 2변수 Muskingum방법의 저류상수 K와 가중계수 x에 추가된 $\alpha$는 측방유입수를 고려해주는 변수이다. 3변수모형의 추적방정식은 유한차분 방정식으로 표현되며, 추적상수 결정은 Matrix Inversion에 의하여 직접 계산가능하며, 이로부터 각홍수사상의 K x $\alpha$값을 구할수 있다. 본 연구를 실유역에 적용하여 실측치와 비교하여본 결과 비교적 잘 맞음을 알 수 있었으며, K와 x값은 하도특성인자로서 홍수규모와 관계되고 측방유입인자 $\alpha$는 항특성에 의하여 지배되는 변수로 측방유입량이 클수록 값이 커지는 성향으로 나타났다.

• Water for future
• /
• v.23 no.3
• /
• pp.385-395
• /
• 1990

Three-parameter Muskingum flood routing model which incorporated the inflows alongside the river channel is applied for the Waegwan-Jeukpogyo reach of the Nakdong River using the flood data of 12 selected flood events experienced in this reach. The flood routing equations for three-parameter model were expressed as a system of finite difference equations and the routing constants were directly computed by matrix inversion method. Then, the three parameters, which consist of the storage constants(K), weighting fator(x), and lateral inflow parameter(α), were determined from the computed routing constants. The results of the present study showed that the model can predict with a fair accuracy the flood discharges at the downsteam end of the reach. The parameters K and x were seen as channel parameters which have close relations with the flood magnitude, whereas the lateral inflow parameter was shown to be strongly governed by the rainfall characteristics of the tributary watersheds contributing to the lateral inflows.

• The Korean Journal of Financial Studies
• /
• v.9 no.1
• /
• pp.245-268
• /
• 2003

전통적인 방법에 의한 PER는 단순히 그 지표의 일반적인 높고 낮음에 따라 소위 '저평가 종목'이라는 이름으로 투자자들에게 추천되고 있다. 그러나 이런 방법은 개별기업의 구체적 내용을 정확하고 종합적으로 고려하지 못하고 있다. 본 연구는 전통적인 방법으로 사용되는 PER지표의 문제점을 개선코자 배당평가 모형으로부터 도출한 PER지표의 구성요소들을 독립변수로 활용 회귀분석을 했다. 그리고 이를 근거로 이론 지표를 만든 후, 그 이론 지표를 투자의사결정에 적용하였을 때의 유효성을 검증했다. PER지표를 구성하는 독립변수는 Kisor & Whitbeck(1963), Malkiel & Cragg(1970), A. Damodaran(1996)에 의해 연구된 것을 원용, PER지표의 구성요소들로 기업의 배당성향, 이익 성장을, 그리고 위험변수로서의 베타계수를 선정했다. 투자성과는 포트폴리오 투자가 일반적인 현실을 감안해 가치가중수익률을 사용한 포트폴리오의 투자성과를 측정했고, 표본은 국내 거래소 시장에 1991년부터 2001년까지 계속 상장된 금융업종을 제외한 전종목을 대상으로 했다. 실증분석에 사용된 기간은 1997년부터 2001년까지 5년 동안의 자료이며, 투자성과를 검증하기 위한 검증모형으로 위험 프리미엄 모형을 사용했다. 먼저 동 분석기간 중 전통적인 방법에 의한 PER효과는 나타나지 않았고, 아울러 기업규모 효과도 찾을 수 없었다. 그러나 회귀분석을 통해 구해진 이론 지표를 활용할 경우, 이론 지표에 비해 시장 지표가 과소 평가된 그룹이 과대 평가된 그룹과 비교할 때 투자성과가 더 우수한 것으로 나타났다. 또한 이론 지표를 통해 PER수준이 낮아짐에 따라 투자성과가 더 높아지는 PER효과도 발견됐다. 이와 같이 이론 지표에 의해 나타나는 PER효과는 기업규모 효과와는 독립적인 것으로 보인다. 외환위기 이후 우리시장에 나타난 차별화 장세 속에 아직도 PER효과나 기업규모 효과와 같은 시장이례 현상이 존재하는지는 관심의 대상이 됐다고 본다. 본 연구에 의하면 기업규모 효과와는 별개의 PER효과가 여전히 존재하며, 다만 이 PER 효과는 전통적 의미의 일반적으로 낮은 PER종목이 초과수익률을 내는 것이 아니라, 기업규모가 크더라도 그 기업의 개별특성을 고려했을 때 이와 비교해 상대적으로 PER가 낮은 종목에 투자하면 초과수익을 낼 수 있음을 의미한다.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.309-327
• /
• 2021

Dynamic treatment regimes (DTRs) are decision-making rules designed to provide personalized treatment to individuals in multi-stage randomized trials. Unlike classical methods, in which all individuals are prescribed the same type of treatment, DTRs prescribe patient-tailored treatments which take into account individual characteristics that may change over time. The Q-learning method, one of regression-based algorithms to figure out optimal treatment rules, becomes more popular as it can be easily implemented. However, the performance of the Q-learning algorithm heavily relies on the correct specification of the Q-function for response, especially in observational studies. In this article, we examine a number of double-robust weighted least-squares estimating methods for Q-learning in high-dimensional settings, where treatment models for propensity score and penalization for sparse estimation are also investigated. We further consider flexible ensemble machine learning methods for the treatment model to achieve double-robustness, so that optimal decision rule can be correctly estimated as long as at least one of the outcome model or treatment model is correct. Extensive simulation studies show that the proposed methods work well with practical sample sizes. The practical utility of the proposed methods is proven with real data example.