• Title/Summary/Keyword: Selection-bias

Search Result 338, Processing Time 0.028 seconds

A Study on the Optimal Discriminant Model Predicting the likelihood of Insolvency for Technology Financing (기술금융을 위한 부실 가능성 예측 최적 판별모형에 대한 연구)

  • Sung, Oong-Hyun
    • Journal of Korea Technology Innovation Society
    • /
    • v.10 no.2
    • /
    • pp.183-205
    • /
    • 2007
  • An investigation was undertaken of the optimal discriminant model for predicting the likelihood of insolvency in advance for medium-sized firms based on the technology evaluation. The explanatory variables included in the discriminant model were selected by both factor analysis and discriminant analysis using stepwise selection method. Five explanatory variables were selected in factor analysis in terms of explanatory ratio and communality. Six explanatory variables were selected in stepwise discriminant analysis. The effectiveness of linear discriminant model and logistic discriminant model were assessed by the criteria of the critical probability and correct classification rate. Result showed that both model had similar correct classification rate and the linear discriminant model was preferred to the logistic discriminant model in terms of criteria of the critical probability In case of the linear discriminant model with critical probability of 0.5, the total-group correct classification rate was 70.4% and correct classification rates of insolvent and solvent groups were 73.4% and 69.5% respectively. Correct classification rate is an estimate of the probability that the estimated discriminant function will correctly classify the present sample. However, the actual correct classification rate is an estimate of the probability that the estimated discriminant function will correctly classify a future observation. Unfortunately, the correct classification rate underestimates the actual correct classification rate because the data set used to estimate the discriminant function is also used to evaluate them. The cross-validation method were used to estimate the bias of the correct classification rate. According to the results the estimated bias were 2.9% and the predicted actual correct classification rate was 67.5%. And a threshold value is set to establish an in-doubt category. Results of linear discriminant model can be applied for the technology financing banks to evaluate the possibility of insolvency and give the ranking of the firms applied.

  • PDF

Dual-Band Frequency Reconfigurable Small Eighth-Mode Substrate-Integrated Waveguide Antenna (이중 대역 주파수 가변 1/8차 소형 기판집적형 도파관 안테나)

  • Kang, Hyunseong;Lim, Sungjoon
    • The Journal of Korean Institute of Electromagnetic Engineering and Science
    • /
    • v.25 no.1
    • /
    • pp.10-18
    • /
    • 2014
  • In this paper, we propose a new frequency reconfigurable dual-band antenna. By using an electronically compact eighth-mode substrate-integrated-waveguide(EMSIW) resonator, we have designed a compact antenna, which performs dual-band movement by additionally loading a complementary split ring resonator(CSRR) structure. The EMSIW and CSRR structures are designed to satisfy the bandwidths of 1.575 GHz(GPS) and 2.4 GHz(WLAN), respectively. We load the CSRR with a varactor diode to allow a narrow bandwidth and to enable the resonance frequency to continuously vary from 2.4 GHz to 2.5 GHz. Thus, we realize a channel selection function that is used in the WLAN standards. Irrespective of how a varactor diode moves, the EMSIW independently resonates so that the antenna maintains a fixed frequency of the GPS bandwidth even at different voltages. Consequently, as the DC bias voltage changes from 11.4 V to 30 V, the resonance frequency of the WLAN bandwidth continuously changes between 2.38 GHz and 2.5 GHz, when the DC bias voltage changes from 11.4 V to 30 V. We observe that the simulated and the measured S-parameter values and radiation patterns are in good agreement with each other.

EM Algorithm and Two Stage Model for Incomplete Data (불완전한 자료에 대한 보완기법(EM 알고리듬과 2단계(Two Stage) 모델))

  • 박경숙
    • Korea journal of population studies
    • /
    • v.21 no.1
    • /
    • pp.162-183
    • /
    • 1998
  • This study examines the sampling bias that may have resulted from the large number of missing observations. Despite well-designed and reliable sampling procedures, the observed sample values in DSFH(Demographic Survey on Changes in Family and Household Structure, Japan) included many missing observations. The head administerd survey method of DSFH resulted in a large number of missing observations regarding characteristics of elderly non-head parents and their children. In addition, the response probability of a particular item in DSFH significantly differs by characteristics of elderly parents and their children. Furthermore, missing observations of many items occurred simultaneously. This complex pattern of missing observations critically limits the ability to produce an unbiased analysis. First, the large number of missing observations is likely to cause a misleading estimate of the standard error. Even worse, the possible dependency of missing observations on their latent values is likely to produce biased estimates of covariates. Two models are employed to solve the possible inference biases. First, EM algorithm is used to infer the missing values based on the knowledge of the association between the observed values and other covariates. Second, a selection model was employed given the suspicion that the probability of missing observations of proximity depends on its unobserved outcome.

  • PDF

A Systematic Review of Moxibustion for Frozen Shoulder (동결견에 대한 뜸치료 효과의 체계적 문헌고찰)

  • Han, Si-Hoon;Lee, Gi-Eon;Byun, Da-Young;Kim, Young-Ik;Kim, Min-Kyun;Min, Boo-Ki;Lim, Han-Bit;Jeong, Yun-Jae;Oh, Min-Seok
    • Journal of Korean Medicine Rehabilitation
    • /
    • v.29 no.3
    • /
    • pp.35-49
    • /
    • 2019
  • Objectives The purpose of this research is to evaluate the effectiveness of moxibustion theraphy for frozen shoulder through systematic review and meta-analysis. Methods The effect of moxibustion for frozen shoulder was constructed as a key question to search randomized controlled trials (RCTs) in 16 databases without type of publication restriction. Results A total of 790 studies were searched and 22 RCTs were finally selected through primary and secondary selection/exclusion process. In most of the evaluation indexes used in each study, moxibustion treatment for shoulder pain was statistically significant compared with other treatments. A meta-analysis of effective rate in 10 studies comparing moxibustion with acupuncture and acupuncture showed a statistically significant effect of moxibustion treatment. And there was a significant effect on the improvement of the shoulder mobility, but the effect size of shoulder pain was no significant effect for shoulder pain reduction. In addition in the result of meta-analysis of 4 studies' effective rate comparing moxibustion with electronic acupuncture, the effect of moxibustion was significantly higher than that of electronic acupuncture in frozen shoulder. Conclusions Moxibustion therapy was effective in symptoms in patients with frozen shoulder. However because of the high risk of bias in the quality of the included studies, the applicability of this study seems to be limited. It is thought that future randomized controlled trials will need to be designed and performed by reviewing ways to minimize bias.

Examining Impact of Weather Factors on Apple Yield (사과생산량에 영향을 미치는 기상요인 분석)

  • Kim, Mi Ri;Kim, Seung Gyu
    • Korean Journal of Agricultural and Forest Meteorology
    • /
    • v.16 no.4
    • /
    • pp.274-284
    • /
    • 2014
  • Crops and varieties are mostly affected by temperature, the amount of precipitation, and duration of sunshine. This study aims to identify the weather factors that directly influence to apple yield among the series of daily measured weather variables during growing seasons. In order to identify them, 1) a priori natural scientific knowledge with respect to the growth stage of apples and 2) pure statistical approaches to minimize bias due to the subject selection of variables are considered. Each result estimated by the Panel regression using fixed/random effect models is evaluated through suitability (i.e., Akaike information criterion and Bayesian information criterion) and predictability (i.e., mean absolute error, root mean square error, mean absolute percentage). The Panel data of apple yield and weather factors are collected from fifteen major producing areas of apples from 2006 to 2013 in Korea for the case study. The result shows that variable selection using factor analysis, which is one of the statistical approaches applied in the analysis, increases predictability and suitability most. It may imply that all the weather factors are important to predict apple yield if statistical problems, such as multicollinearity and lower degree of freedom due to too many explanatory variables used in the regression, can be controlled effectively. This may be because whole growth stages, such as germination, florescence, fruit setting, fatting, ripening, coloring, and harvesting, are affected by weather.

A Study on Alcohol Expectancy of Elementary Schoolchild (초등학생들의 음주기대에 관한 연구)

  • Lim, Mi-Suk;Park, Young-Soo
    • The Journal of Korean Society for School & Community Health Education
    • /
    • v.3
    • /
    • pp.15-33
    • /
    • 2002
  • Researchers' common findings is that there are positive or negative effect of alcohol expectancy on drinking behavior. Therefore we would effectively prevent troublesome drinking of the youth and university students by inquiring and controlling critical factors affecting alcohol expectancy. The purposes of this thesis are, first, to empirically test factors affecting the alcohol expectancy level of elementary schoolchild(potential drinker).; second, to suggest the necessity for development of pre-alcohol prevention programs. On the basis of previous research, eighteen factors included in four categories(general characters, environmental characters, alcohol knowledge, drinking experience) affecting alcohol expectancy level were found out. 623 subjects used in this study were drawn from 8 elementary schools in Daegu, Korea. The empirical results suggested that the alcohol expectancy level of elementary schoolchildren was negative in general. And it was proved that 9 factors were significantly correlated with alcohol expectancy level. To put it concretely(see Fig.), (1) It was proved that schoolchildren with bad environment(live in oneself, displeased drinking feeling) rather than good environment(live with parents, nice drinking feeling) for drinking had more negative alcohol expectancy. (2) Korean traditional culture that partakes of sacrificial food and drink have an influence on the first drinking of most elementary schoolchildren. And it was proved that schoolchildren with this drinking experience rather than any other motives had less negative alcohol expectancy. (3) It was proved that schoolchildren adapting themselves rather than being difficult in school life had more negative alcohol expectancy. And the more knowledge about alcohol or drinking schoolchildren had, the more they had negative alcohol expectancy (4) It was proved that schoolchildren having drinking experience or drinking at present rather than having no drinking experience or not-drinking at present had less negative alcohol expectancy. (5) It was proved that schoolchildren having strong drinking intention rather than having weak or no drinking intention in the future had more positive alcohol expectancy. Based on previous results, guideline for development of pre-alcohol prevention programs can be represented: discriminated programs development on educatee, drinking education programs development increasing the power of self-control about alcohol and drinking, social education or continuing education programs development on drinking, open preschool education to substantially prevent drinking or alcoholism etc. The findings, however, should be interpreted with caution, because this study has several limitations in measurement and sampling as follows. First, selection bias because of limited selection of sampling. It is because the subjects are drawn from only 8 elementary schools in Daegu. Second, less refined measurement ; Therefore, it is necessary to develop more detailed measures on alcohol knowledge, alcohol expectancy level especially. Further researches should be suggested and encouraged with more refined methodologies.

  • PDF

The Effect of Auricular Acupuncture for Obesity: A Systematic Review and Meta-analysis (비만치료에 대한 이침요법의 효과에 대한 고찰: 체계적 문헌고찰 및 메타분석)

  • Park, Seohyun;An, Sunjoo;Choi, Sunghwan;Kang, Shinwoo;Keum, Dongho
    • Journal of Korean Medicine for Obesity Research
    • /
    • v.20 no.1
    • /
    • pp.52-68
    • /
    • 2020
  • Objectives: The purpose of this study is to investigate the weight loss effects of auricular acupuncture in obese patients. Methods: Six databases (Research Information Sharing Service [RISS], Korean studies Information Service System [KISS], Oriental Medicine Advanced Searching Integrated System [OASIS], PubMed, The Cochrane Library, China National Knowledge Infrastructure [CNKI]) were searched up to May 20, 2020. Eight eligible randomized controlled trials were included the present study. The quality of included studies was assessed by the Cochrane risk of bias tool and a meta-analysis was performed by Review Manager software. A meta-analysis was conducted using a random-effects model and a subgroup analysis was performed to detect the sources of heterogeneity, identify the selection of acupuncture methods and explore its contributions to the weight loss effects. Results: Among 8 trials, 5 trials used auricular acupuncture and 2 trials used auricular acupressure, 1 trial used both types of intervention. Most commonly selected acupoints were Shenmen (TF4) and stomach (CO4). Treatment duration was six to twelve weeks, and total treatment session was six to twelve. Compared to the control groups, auricular acupuncture significantly decreased body weight, body mass index (BMI), high density lipoprotein-cholesterol (HDL) and ghrelin. For the selection of acupuncture methods, both methods performed similarly in most outcome except waist circumference (WC), body fat percentage (BFP), and triglycerides (TG). Conclusions: We found that auricular acupuncture can be effective for weight loss and controlling appetite. However, the findings should be interpreted with caution due to heterogeneity. So further vigorous and well-designed studies should be conducted to strengthen the evidence of the use of auricular acupuncture for obesity.

Does performing high- or low-risk coronary artery bypass graft surgery bias the assessment of risk-adjusted mortality rates of hospitals? (관상동맥우회로술의 위험 수준이 병원내사망률 평가 결과에 미친 영향 분석)

  • Lee, Kwang-Soo;Lee, Sang-Il;Lee, Jung-Soo
    • Health Policy and Management
    • /
    • v.17 no.3
    • /
    • pp.87-105
    • /
    • 2007
  • The purpose of this study was to analyze whether nonemergency, isolated coronary artery bypass graft (CABG) surgery for high- or low-risk patients biases the assessment of the risk-adjusted mortality rates of hospitals. This study used 2002 National Health Insurance claims data for tertiary hospitals in Korea. The study sample consisted of 1,959 patients from 23 tertiary hospitals. The risk-adjustment model used the patients' biological, admission, and comorbidity data identified in the claims. The subjects were classified into high- and low-risk groups based on predicted surgical risk. The crude mortality rates and risk-adjusted mortality rates for low-risk, high-risk, and all patients in a hospital were compared based on the rank and the four intervals defined by quartile. Also, the crude mortality rates of the three groups were compared with their 95% confidence intervals of predicted mortality rates. The C-statistic (0.83) and Hosmer-Lemeshow test ($X^2$=11.47, p=0.18) indicated that the risk-adjustment model performed well. Presenting crude mortality rates with their 95% confidence intervals of predicted rates showed higher agreements among the three groups than using the rank or intervals of mortality rates defined by quartile in the hospital performance assessment. The crude mortality rates for the low-risk patients in 21 of the 23 hospitals were located on the same side of their 95% confidence intervals compared to that for all patients. High-risk patients and all patients differed at only one hospital. In conclusion, the impact of risk selection by hospital on the assessment results was the smallest when comparing the crude inpatient mortality rates of CABG patients with the 95% confidence intervals of predicted mortality rates. Given the increasing importance of quality improvements in Korean health policy, it will be necessary to use the appropriate method of releasing the hospital performance data to the public to minimize any unwanted impact such as risk-based hospital selection.

User-Centered Climate Change Scenarios Technique Development and Application of Korean Peninsula (사용자 중심의 기후변화 시나리오 상세화 기법 개발 및 한반도 적용)

  • Cho, Jaepil;Jung, Imgook;Cho, Wonil;Hwang, Syewoon
    • Journal of Climate Change Research
    • /
    • v.9 no.1
    • /
    • pp.13-29
    • /
    • 2018
  • This study presented evaluation procedure for selecting appropriate GCMs and downscaling method by focusing on the climate extreme indices suitable for climate change adaptation. The procedure includes six stages of processes as follows: 1) exclusion of unsuitable GCM through raw GCM analysis before bias correction; 2) calculation of the climate extreme indices and selection of downscaling method by evaluating reproducibility for the past and distortion rate for the future period; 3) selection of downscaling method based on evaluation of reproducibility of spatial correlation among weather stations; and 4) MME calculation using weight factors and evaluation of uncertainty range depending on number of GCMs. The presented procedure was applied to 60 weather stations where there are observed data for the past 30 year period on Korea Peninsula. First, 22 GCMs were selected through the evaluation of the spatio-temporal reproducibility of 29 GCMs. Between Simple Quantile Mapping (SQM) and Spatial Disaggregation Quantile Delta Mapping (SDQDM) methods, SQM was selected based on the reproducibility of 27 climate extreme indices for the past and reproducibility evaluation of spatial correlation in precipitation and temperature. Total precipitation (prcptot) and annual 1-day maximum precipitation (rx1day), which is respectively related to water supply and floods, were selected and MME-based future projections were estimated for near-future (2010-2039), the mid-future (2040-2069), and the far-future (2070-2099) based on the weight factors by GCM. The prcptot and rx1day increased as time goes farther from the near-future to the far-future and RCP 8.5 showed a higher rate of increase in both indices compared to RCP 4.5 scenario. It was also found that use of 20 GCM out of 22 explains 80% of the overall variation in all combinations of RCP scenarios and future periods. The result of this study is an example of an application in Korea Peninsula and APCC Integrated Modeling Solution (AIMS) can be utilized in various areas and fields if users want to apply the proposed procedure directly to a target area.

Taegeuk Acupuncture for Patients with Chronic Musculoskeletal Pain: A Retrospective Analysis of Medical Records (만성 근골격계 통증환자에 대한 태극침법의 임상적 유효성: 후향적 의무기록 분석 연구)

  • Kim, Jae Kyu;Kim, Kun Hyung;Noh, Seung Hee;Kim, Yu Ri
    • Journal of Acupuncture Research
    • /
    • v.31 no.2
    • /
    • pp.145-152
    • /
    • 2014
  • Objectives : This study aimed to assess the effectiveness and safety of Taegeuk acupuncture for patients with chronic musculoskeletal pain. Methods : A retrospective analysis of the electronic medical records of Pusan National University Korean Medicine Hospital from March 2012 to March 2013 was performed. Eligible cases were outpatients who had received at least six sessions of Taegeuk acupuncture for their pain and had agreed to the use of their medical records for research purposes. Pain levels, heart rate variability, and patients' perceptions of the safety of acupuncture were investigated through the medical records and follow-up telephone interviews. Results : Fifteen cases were selected for analysis. All had suffered from musculoskeletal pain for at least one year. After six sessions of Taegeuk acupuncture, the average score on the pain numeric rating scale decreased from $6.7{\pm}2.0$ to $2.9{\pm}2.2$(a 43 % reduction). Follow-up telephone interviews revealed that patients experienced no adverse event after Taegeuk acupuncture. Conclusions : Patients with chronic musculoskeletal pain showed a reduction of more than half of the baseline pain after six sessions of Taegeuk acupuncture. However, the beneficial effects observed in this retrospective analysis should be interpreted with caution due to selection and recall bias. Further randomized controlled trials of Taegeuk acupuncture for chronic musculoskeletal pain are warranted.