• Title/Summary/Keyword: Probit estimation

Search Result 57, Processing Time 0.028 seconds

A Bayesian inference for fixed effect panel probit model

  • Lee, Seung-Chun
    • Communications for Statistical Applications and Methods
    • /
    • v.23 no.2
    • /
    • pp.179-187
    • /
    • 2016
  • 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.

On a Bayes Criterion for the Goodness-of-Link Test for Binary Response Regression Models : Probit Link versus Logit Link

  • Kim, Hea-Jung
    • Journal of the Korean Statistical Society
    • /
    • v.26 no.2
    • /
    • pp.261-276
    • /
    • 1997
  • In the context of binary response regression, the problem of constructing Bayesian goodness-of-link test for testing logit link versus probit link is considered. Based upon the well known facts that cdf of logistic variate .approx. cdf of $t_{8}$/.634 and, as .nu. .to. .infty., cdf of $t_{\nu}$ approximates to that of N(0,1), Bayes factor is derived as a test criterion. A synthesis of the Gibbs sampling and a marginal likelihood estimation scheme is also proposed to compute the Bayes factor. Performance of the test is investigated via Monte Carlo study. The new test is also illustrated with an empirical data example.e.

  • PDF

A Study on Estimation of Human Damage for Overpressure by Vapor Cloud Explosion in Enclosure Using Probit Model (프로빗모델을 통한 밀폐공간에서의 증기운폭발 과압에 의한 인체피해예측)

  • Leem, Sa-Hwan;Lee, Jong-Rark;Huh, Yong-Jeong
    • Journal of the Korean Institute of Gas
    • /
    • v.12 no.1
    • /
    • pp.42-47
    • /
    • 2008
  • The demand of gas as an eco-friendly energy source has being increased. With the demand of gas, the use of gas is also increased, so injury and loss of life by the explosion and fire have been increasing every year. Hence the influence on over-pressure caused by Vapor Cloud Explosion in enclosure of experimental booth was calculated by using the Hopkinson's scaling law and damage effect by the accident to a human body was estimated by applying the probit model. As a result of the damage estimation conducted by using the probit model, both the damage possibility of explosion overpressure to human over 3 meters away and that of overpressure to tympanum rupture over 25 meters away from the explosion shows nothing.

  • PDF

A Study on Estimation of Overpressure Damage Caused by Rupture of Butane Can (volume : 34 g) (부탄 캔(용량 : 34 g)파열로 인한 과압의 피해예측에 관한 연구)

  • Leem Sa Hwan;Choi Ic Whoan;Lim Dong Yeon
    • Journal of the Korean Institute of Gas
    • /
    • v.9 no.2 s.27
    • /
    • pp.8-15
    • /
    • 2005
  • With the introduction of 40 hour working week system, more households enjoy picnics on weekends. More gas accidents take place on Saturdays and on Sundays than any other days of week. As of October, 2004 casualties resulted from butane can accidents increased 1.5 times compared to the same period of the previous year. In this study, the influence of explosion over-pressure caused by the rupture of butane can thrown away after use was calculated by using the Hopkinson's Scaling Law and the accident damage was estimated by applying the influence on the adjacent structures and people into the Probit model. As a result of the damage estimation conducted by using the Probit model, both the damage possibility of explosion over-pressure to structures 50 meters away and that of over-pressure to people 10 meters away showed nothing. The explosion efficiency used was 100 percent. As a result of this, the actual damage influenced by the rupture of butane can would be lower than the value calculated in this study and expected to be safer.

  • PDF

Innovation and FDI: Applying Random Parameters Methods to KIS Data (기술혁신과 FDI)

  • Kim, Byung-Woo
    • Journal of Korea Technology Innovation Society
    • /
    • v.13 no.3
    • /
    • pp.513-537
    • /
    • 2010
  • According to the "FDI-as-market-discipline" hypothesis, inward FDI acts as a mechanism of change in market structure affecting innovative activities of domestic firms. We used panel KIS data for testing this hypothesis. Binary probit estimation shows that, in contrast to the German case of Bertschek (1995), FDI is insignificant in Korean case for explaining product innovation. 1his result maybe comes from the fact that the industries in Korea are more monopolistic or oligopolistic than those of Germany. Using panel data, we tried random parameter estimation using matrix weighted average of GLS and OLS. The result shows different estimates from cross-section outcome and panel estimation with parameter homogeneity, so we can infer large parameter heterogeneity across firms. But, interpretation for FDI variable is similar across panel and cross-section estimation.

  • PDF

A Study on the Estimation of Human Damage Caused by Vapor Cloud Explosion(VCE) in LPG Filling Station (LPG자동차충전소에서 증기운폭발로 인한 인명피해예측에 관한 연구)

  • Leem, Sa-Hwan;Huh, Yong-Jeong
    • Journal of the Korean Institute of Gas
    • /
    • v.14 no.2
    • /
    • pp.15-21
    • /
    • 2010
  • The demand of gas as an eco-friendly energy source has being increased. With increasing the LPG demand, the number of LPG filling station. In this work, the influence on over-pressure caused by Vapor Cloud Explosion in gas station was calculated by using the Hopkinson's scaling law and injury effect by the accident to a human body was estimated by applying the probit model. As a result of the injury estimation conducted by using the probit model for leakage 10% of 20ton storage tank. The distances from LPG station for death and tympanum rupture are 36.5 and 290 meters, respectively.

An Analysis of the Amount of Willingness to Pay for Environment-friendly Agricultural Processed Foods by the Level of Consumption Intention (친환경농산물가공식품 소비의향 수준별 지불의향금액 분석)

  • Jeong, Hak-Kyun;Han, Jae-Hwan
    • Korean Journal of Organic Agriculture
    • /
    • v.26 no.4
    • /
    • pp.543-558
    • /
    • 2018
  • The purpose of this study is to analyze the amount of willingness to pay for environment-friendly agricultural processed foods by the level of consumption intention. To accomplish the objective of the study a consumer survey was conducted for quantitative analysis regarding consumption pattern. The ordered-probit selection model, Heckman-type two-stage method was employed for an empirical analysis on determining the amount of willingness to pay. The estimation results showed that in the case of consumers who are to keep their consumption at the present level, those with high quality contentment compared to price, brand contentment, and purchasing chiefly in the special store, have the higher amount of willingness. The estimation results also showed that in the case of consumers who are to increase their consumption, those with high quality contentment compared to price, brand contentment, certificate contentment, and having family members who suffer from diseases such as atopy, cancer, diabetes, high blood pressure, have the higher amount of willingness. Therefore For those who are to increase their consumption, enhancement of quality, enhancement of contentment with certificate, promotion are useful to increase the amount of willingness to pay.

The Health Effects of PM2.5: Evidence from Korea (대기오염의 건강위해성 연구 - PM2.5를 중심으로 -)

  • Hong, Jong-Ho;Ko, Yookyung
    • Environmental and Resource Economics Review
    • /
    • v.12 no.3
    • /
    • pp.469-485
    • /
    • 2003
  • This paper reports on the results of epidemiological investigation of daily health effects in the elderly associated with daily exposure to particulate matters in Korea. Our main focus is on the potential difference in health effects between PM10 and PM2.5. While the Korean environmental authority has set an ambient standard for PM10, the government currently does not monitor PM2.5, which has no national standard. A daily data on respiratory symptoms as well as PM concentrations are collected for a total of 120 days. Using a probit model, we find statistically significant negative health effects of PM2.5 on respiratory symptoms among the nonsmoking elderly, while PM10 does not show such effects from the estimation. This result suggests that, for air quality regulatory purposes, PM2.5 can be a more appropriate air pollutant than PM10.

  • PDF

A Comparative Study on the Method of Consequence Estimation for Release of Toxicant Substances (독성물질 누출의 강도 산정 방법에 관한 비교 연구)

  • 김윤화;백종배;고재욱
    • Journal of the Korean Society of Safety
    • /
    • v.9 no.1
    • /
    • pp.89-94
    • /
    • 1994
  • Two methods, the numerical method of CPQRA and the manual method of IAEA, were used to estimate the effect distance from release and dispersion of toxic materials. The Gaussian plume model which has a weather stability class D with wind velocity of 5m/s was applied to calculate dispersion of toxic materials. Also, probit function were employed to evaluate the human fatality as a result of exposure to toxic gases. Furthermore, concentration of toxic materials corresponding to LC$_{50}$ for 30 min could be determined by setting Pr as 5.0 and solving the probit function. Calculations were conducted by employing chlorine and ammonia as toxic materials because they are not only most commonly used In chemical plants but also very harmful to humans. Calculated results by employing toxic materials indicated that the effect distance from the CPQRA method was between the minimum and maximum distance from the method proposed by IAEA.A.

  • PDF

A Study on the Quantitative Analysis for Explosion of LPG Storage Tank (LPG 저장탱크의 폭발에 대한 정량적 영향평가에 관한 연구)

  • Leem, Sa Hwan;Huh, Yong Jeong
    • Journal of the Korean Institute of Gas
    • /
    • v.17 no.3
    • /
    • pp.1-7
    • /
    • 2013
  • The influence of the over-pressure caused by Explosion in gas station was calculated by using the Hopkinson's scaling law and injury effect by accident to buildings and human bodies was estimated by applying the probit model. As a result, the injury estimation was conducted by using the probit model for leakage 10% of 20ton storage tank. The separate distances from LPG station for building(damage) and human(lung hemorrhage to death) are 260 and 30 meters, respectively.