• 응용통계연구
• /
• 제16권2호
• /
• pp.351-364
• /
• 2003

본 논문에서는 오차성분을 가지는 패널회귀모형에서 모형의 선형성을 검정 할 수 있는 검 정통계량을 제시하고, 유도한 검정통계량의 계산을 위하여 인공회귀방법을 이용하려한다. 모의실험 결과, Double-Length Artificial Resression(DLR)을 이용한 LM 검정통계량은 명목유의 수준을 잘 유지하고 있는 것으로 나타났으며 검정력에 있어서도 기존의 검정에 비하여 높게 나타났다.

• 응용통계연구
• /
• 제34권5호
• /
• pp.697-710
• /
• 2021

상대오차를 이용한 예측법은 상대오차(혹은 퍼센트오차)가 중요시되는 분야, 특히 계량경제학이나 소프트웨어 엔지니어링, 또는 정부기관 공식통계 부분에서 기존 예측방법 외에 선호되는 예측방법이다. 그 동안 상대오차를 이용한 예측법은 선형 혹은 비선형 회귀분석 뿐 아니라, 커널회귀를 이용한 비모수 회귀모형, 그리고 정상시계열분석에 이르기까지 그 범위가 확장되어 왔다. 그러나, 지금까지의 분석은 고정효과(fixed effect)만을 고려한 것이어서 임의효과(random effect)에 관한 상대오차 예측법에 대한 확장이 필요하였다. 본 논문의 목적은 상대오차예측법을 일반화선형혼합모형(GLMM)에 속한 감마회귀(gamma regression), 로그정규회귀(lognormal regression), 그리고 역가우스회귀(inverse gaussian regression)의 패널자료(panel data)에 적용시키는데 있다. 이를 위해 실제 자동차 보험회사의 손해액 자료를 사용하였고, 최량예측량과 최량상대오차예측량을 각각 적용-비교해 보았다.

• 통상정보연구
• /
• 제12권4호
• /
• pp.151-170
• /
• 2010

The present study made an empirical analysis for investigating the competitiveness of technology trades in Korea. In particular, the study deduced the correlation between technology export and technology import using the variables of Gross Domestic Expenditure on R&D and Per capita industry value added Productivity and employed fixed effect model in panel linear regression model. It is found that the R&D expenditure of OECD countries made a significant effect on the technology import and the value-added labor productivity made a significant result on both technology export and import. Therefore, it showed that the technology trade in Korea made a sensitive response to labor productivity in OECD countries. By panel analysis, machine, construction, ICT, and service industry affect most on technology export in Korea for recent 5 years. For technology import, electric-electron, chemical, service, and construction industry have significant effects. This study contributed to understanding of industrial characteristics affecting technology trades in Korea and empirical analysis to show correlation between the factors affecting technology trade.

• Communications for Statistical Applications and Methods
• /
• 제26권3호
• /
• pp.315-323
• /
• 2019

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.

• 유통과학연구
• /
• 제12권11호
• /
• pp.25-34
• /
• 2014

Purpose - Considering that the governments' official statistics on the optimum scale of the domestic service industry will be crucial in future, this study's results will be used as an important benchmark to develop and verify the parameters in the government's official statistics. Research design, data, and methodology - To identify the appropriate scale of Korea's service industry and its adequacy, I have determined them through estimation using a regression method involving panel data analysis on the panel data of 30 OECD countries. Results - The regression coefficient provided indications of being non-linear. This means that a U-shaped curve relationship exists-that is, the level of the economic growth leverage decreases along with the service industry's growth up to the level of 70.9% in terms of the Korean service industry's adequacy; it increases along with the service industry's growth at a level higher than 70.9%. Conclusions - While the current proportion of the size of the service industry among all industries in Korea stands at 50.7%, its proper proportion estimated by a regression analysis was 70.9%.

• Journal of the Korean Data and Information Science Society
• /
• 제28권3호
• /
• pp.547-557
• /
• 2017

건강 관련 삶의 질 자료는 정규분포를 따르지 않고 치우친 분포를 보이며, 등분산 가정을 만족하지 않는 경우가 대부분이다. 또한 건강 관련 삶의 질 자료는 범위가 정해져 있는 자료이며, 건강한 상태를 나타내는 경우 최대값을 가지는 천장효과가 있는 자료이다. 본 연구에서는 건강 관련 삶의 질 자료인 EQ-5D에 대해 선형회귀모형과 베타회귀모형, 그리고 평균과 정밀도에 대한 하위모형을 가지고 있는 확장된 베타회귀모형을 이용하여 예측모형을 개발하고 모형의 예측 정확도를 비교하였다. 선형회귀모형에 비해 확장된 베타회귀모형의 예측 정확도가 높기는 하지만 신뢰구간이 겹치고 있기 때문에 확장된 베타회귀모형의 정확도가 더 높다고 할 수는 없다. 하지만 확장된 베타회귀모형은 공변량에 따라 분산이 달라지는 부분을 설명할 수 있으며 선형회귀모형이 제한된 범위를 벗어난 값을 예측하는 부분을 개선할 수 있다. 따라서 범위가 제한되고 이분산이 있는 치우친 자료에 대해 공변량들이 평균 및 정밀도에 영향을 주는 정도를 동시에 고려하는 확장된 베타회귀모형은 건강 관련 삶의 질 자료인 EQ-5D를 분석하는 방법으로 적절하다고 할 수 있다.

• Asian-Australasian Journal of Animal Sciences
• /
• 제8권5호
• /
• pp.505-513
• /
• 1995

The major factors affecting productivity on daily farms in Queensland, Australia, were determined using the stepwise linear regression approach. The data were obtained from a survey conducted on the total population of daily farms in Queensland in 1987. These data were divided into six major dailying regions. The technique was applied using 12 independent variables believed by a panel of experienced research and extension personnel to exert the most influence on milk production. The regression equations were all significant (p < 0.001) with the percentage coefficients of determination ranging from 62 to 76% for equations developed using' total farm milk: production as the dependent variable. Three of the variables affecting total farm milk: production were found to be common to all six regions. These were; the amount of supplementary energy fed, the area set aside to irrigate winter feed and the size of the area used for dailying. Higher production farms appeared to be more efficient in that they consistently produced milk production levels higher than those estimated from the regression equation for their region. Other methods of analysis including robust regression and non linear regression techniques were unsuccessful in overcoming this problem and allowing development of a model appropriate for farms at all levels of production.

• The Journal of Asian Finance, Economics and Business
• /
• 제9권2호
• /
• pp.363-370
• /
• 2022

The study aims to probe the impact of foreign ownership on Vietnamese listed firms' capital structure. This study employs panel data of 288 non-financial firms listed on the Ho Chi Minh City stock exchange (HOSE) and Ha Noi stock exchange (HNX) in 2015-2019. In this research, we applied a Bayesian linear regression method to provide probabilistic explanations of the model uncertainty and effect of foreign ownership on the capital structure of non-financial listed enterprises in Vietnam. The findings of experimental analysis by Bayesian linear regression method through Markov chain Monte Carlo (MCMC) technique combined with Gibbs sampler suggest that foreign ownership has substantial adverse effects on the firms' capital structure. Our findings also indicate that a firm's size, age, and growth opportunities all have a strong positive and significant effect on its debt ratio. We found that the firms' profitability, tangible assets, and liquidity negatively and strongly affect firms' capital structure. Meanwhile, there is a low negative impact of dividends and inflation on the debt ratio. This research has ramifications for business managers since it improves a company's financial resources by developing a strong capital structure and considering foreign investment as a source of funding.

• 한국수산과학회지
• /
• 제21권3호
• /
• pp.169-176
• /
• 1988

In order to obtain the objective indices which can assess the quality and the shelf-life of dried squid, nonenzymatic browning, carbonyl value, trimethyl amine (TMA) and trimethyl amine oxide (TMAO) decrease and panel test were determined in dried squid at various water activity levels. When the data of nonenzymatic browning fit a zero order equation, $r^2$ value were more than 0.92 except aw 0.52 for 0.8241. Through variance analysis for the data of browning extent and TMAO decrease, the confidence limits of regression equation were $99\%$ and their limit values of shelf-life were shown 0.45390 O. D./g. solid and 190.322 g/g respectively. In case of TMAO, $r^2$ value was calculated more than 0.95. Linear regression equation for the correlation between browning data and average panel score was Y=0.6138-0.053X and its $r^2$ value was 0.9285. Also in TMAO decrease, the equation was InY=2.0314+0.08269x and $r^2$ value was 0.7854. The shelf-life, evaluated by nonenzymatic browning, TMAO decrease and panel test, was 110-170 days at aw 0.45-0.76 except aw 0.15.

• Child Studies in Asia-Pacific Contexts
• /
• 제5권1호
• /
• pp.51-62
• /
• 2015

This study examined the role of ethnicity and extended family relationship in the longitudinal patterns of unmarried resident fathers' engagement with young children in the U.S. Using three waves of panel data from the Fragile Families and Child Wellbeing Study (FFCWS), data obtained from 292 respondents were analyzed by the Linear Mixed Effects Regression (LMER) method. Findings suggested that Black unmarried resident fathers exhibited more engagement with their infants at age one than their White counterparts did, but the initial difference reversed by age five as a result of decreasing engagement among Black fathers over time and comparatively increasing engagement among Whites. Results also suggested that Black unmarried resident fathers with a high relationship level with their extended family members exhibited less engagement with their infants than Black fathers with a low relationship level, but the initial difference reversed by age five as a result of decreasing engagement among Black fathers with a low relationship level over time and comparatively increasing engagement among Black fathers with a high relationship level.