• Journal of Korea Water Resources Association
• /
• v.55 no.10
• /
• pp.723-736
• /
• 2022

In this study, monthly precipitation forecasting models that can predict up to 12 months in advance were constructed for the Geum River basin, and two statistical techniques, multiple linear regression (MLR) and artificial neural network (ANN), were applied to the model construction. As predictor candidates, a total of 47 climate indices were used, including 39 global climate patterns provided by the National Oceanic and Atmospheric Administration (NOAA) and 8 meteorological factors for the basin. Forecast models were constructed by using climate indices with high correlation by analyzing the teleconnection between the monthly precipitation and each climate index for the past 40 years based on the forecast month. In the goodness-of-fit test results for the average value of forecasts of each month for 1991 to 2021, the MLR models showed -3.3 to -0.1% for the percent bias (PBIAS), 0.45 to 0.50 for the Nash-Sutcliffe efficiency (NSE), and 0.69 to 0.70 for the Pearson correlation coefficient (r), whereas, the ANN models showed PBIAS -5.0~+0.5%, NSE 0.35~0.47, and r 0.64~0.70. The mean values predicted by the MLR models were found to be closer to the observation than the ANN models. The probability of including observations within the forecast range for each month was 57.5 to 83.6% (average 72.9%) for the MLR models, and 71.5 to 88.7% (average 81.1%) for the ANN models, indicating that the ANN models showed better results. The tercile probability by month was 25.9 to 41.9% (average 34.6%) for the MLR models, and 30.3 to 39.1% (average 34.7%) for the ANN models. Both models showed long-term predictability of monthly precipitation with an average of 33.3% or more in tercile probability. In conclusion, the difference in predictability between the two models was found to be relatively small. However, when judging from the hit rate for the prediction range or the tercile probability, the monthly deviation for predictability was found to be relatively small for the ANN models.

• Journal of Chest Surgery
• /
• v.39 no.8 s.265
• /
• pp.579-587
• /
• 2006

Background: Preoperative risk analysis for Fontan candidates is still less than optimal in that patients with apparently low risks may have poor surgical outcome; prolonged pleural drainage, protein losing enteropathy, pulmonary thromboembolism and death. We hypothesized that low pulmonary vascular compliance (PVC) is a risk factor for prolonged pleural effusion drainage after the Fontan operation. Material and Method: A retrospective review of 96 consecutive patients who underwent the Extracardiac Fontan procedures (median age: 3.9 years) was performed. Fontan risk score (FRS) was calculated from 12 categorized preoperative anatomic and physiologic variables. PVC $(mm^2/m^2{\cdot}mmHg)$ was defined as pulmonary artery index $(mm^2/m^2)$ divided by total pulmonary resistance $(W.U{\cdot}/m^2)$ and pulmonary blood flow $(L/min/m^2)$ based on the electrical circuit analogue of the pulmonary circulation. Chest tube indwelling time was log-transformed (log indwelling time, LIT) to fit normal distribution, and the relationship between preoperative predictors and LIT was analyzed by multiple linear regression. Result: Preoperative PVC, chest tube indwelling time and LIT ranged from 6 to 94.8 $mm^2/mmHg/m^2$ (median: 24.8), 3 to 268 days (median: 20 days), and 1.1 to 5.6 (mean: 2.9, standard deviation: 0.8), respectively. FRS, PVC, cardiopulmonary bypass time (CPB) and central venous pressure at postoperative 12 hours were correlated with LIT by univariable analyses. By multiple linear regression, PVC (p=0.0018) and CPB (p=0.0024) independently predicted LIT, explaining 21.7% of the variation. The regression equation was LIT=2.74-0.0158 PVC+0.00658 CPB. Conclusion: Low pulmonary vascular compliance is an important risk factor for prolonged pleural effusion drainage after the extracardiac Fontan procedure.

• Clinical and Experimental Pediatrics
• /
• v.48 no.1
• /
• pp.75-80
• /
• 2005

Purpose : Internal tibial torsion is prevalent in East Asian countries such as Korea and Japan, where sitting on the floor is common behavior. Internal tibial torsion or excessive lateral tibial torsion may cause esthetical, functional, or psychological problems and also may induce degenerative arthritis in older age. The purpose of this study is to measure the tibial torsion in children of the Jeju area. Methods : Tibial torsion was measured in 1,042 lower extremities of 521 children from one to 12 years of age. The values of transmalleolar angles were analyzed for each age group divided by 6 months. Quadratic and linear regression models were used to fit patterns of changes in mean values of transmalleolar angles. The age at seven, which provides the highest coefficient of determination for quadratic regression analysis, was used as a cut-off point to fit different statistical models. Results : The mean transmalleolar angle was $0.10{\pm}5.79^{\circ}$ in all children,$ 0.90{\pm}5.49^{\circ}$ in males, and $-0.80{\pm}5.97^{\circ}$ in females. The value was $4.25{\pm}4.04$ in 1 year of age, gradually decreased to the lowest level of $-1.98^{\circ}$ in four years and seven months of age, increased again with age until it reached $0.67{\pm}1.10^{\circ}$ at seven years of age, and stayed at that level thereafter. Conclusion : Internal tibial torsion in infancy is known to correct spontaneously in the normal developing process. But in this study, the mean transmalleolar angle in children of Jeju area annually decreased after one year of age; to the lowest angle at four years and seven months of age; increased again gradually to the age of seven; and persisted in that level, about $10^{\circ}$ less than western children, not correcting further thereafter. These findings suggest tibial torsion might be caused by lifestyle, especially sitting on feet. To prevent abnormalities of joints and gaits, early diagnosis of tibial torsion in childhood and posture correction or early treatment when needed, seems to be necessary.

• 한국사회복지학회:학술대회논문집
• /
• 2003.10a
• /
• pp.185-216
• /
• 2003

This study was aimed at investigating important factors influencing career success among regular workers. The current researcher scrutinized the degree to which variables and factors affect the career success and occupational turnover rates of the research participants. At the same tune, two hypothetical path models established by the researcher were examined using linear multiple regression methods and the LISREL. After examining the differences among the factors of career success, a comparison was made between the disabled worker group and the non-disabled worker group. A questionnaire using the 5-point Likert scale was distributed to a group of 374 workers with disabilities and 463 workers without disabilities. For the data analysis purpose, the structural equation model, factor analysis, correlation analysis, and multiple regression analysis were carried out. The results of this study ran be summarized as follows. First, the results of factor analysis showed important categories of conceptual themes of career success. The initial conceptual factor model did not accord with the empirical one. A three-factorial model revealed categories of personal, family, and organizational factor respectively. The personal factor was composed of the self-esteem and self-efficiency. The family factor was consisted of the multi-roles stress and the number of children. Finally, the organizational factor was composed of the capacity for utilizing resources, networking, and the frequency of mentoring. In addition, the total 10 sub areas of career success were divided by two important aspects; the subjective career success and the objective career success. Second, both research participant groups seemed to be influenced by their occupational types. However, all predictive variables excluding the wage rate and the average length of work years had significant impact on job success for the disabled work group, while all the variables excluding the frequency of advice and length of working years had significant impact on job success for the non-disabled worker group. Third, the turnover rate was significantly influenced by the age and the experience of turnover of the research participants. However, the number of co-workers was the strongest predictive variable for the worker group with disabilities, but the occupation choice variable for the worker group without disabilities. For the disabled worker group, the turnover rate was differently influenced by the type of occupation, the length of working years, while multi-role stress and the average working years at the time of turnover for the worker group without disabilities. Fifth, as a result of verifying the hypothetical path model, it showed that the first model was somewhat proper and could predict the career success on both research participant groups. In the second model, the Chi-square, the degree of freedom (($x^2=64.950$, df=61, P=0.341), and the adjusted Goodness of Fit Index (AGFI) were .964, and the Comparative Fit Index (CFI) were .997, and the Root Mean Squared Residual (RMR) was respectively. .038. The model was best fitted and could predict the career success more highly because the goodness of fit index in the whole models was within the allowed range. In conclusion, the following research implications can be suggested. First, the occupational type of research participants was one of the most important variables to predict the career success for both research participant groups. It means that people with disabilities require human development services including education. They need to improve themselves in this knowledge-based society. Furthermore, for maintaining the career success, people with disabilities should be approached by considering the subjective career success aspects including wages and the promotion opportunities than the objective career success aspects.

• Journal of the Korea Organic Resources Recycling Association
• /
• v.3 no.2
• /
• pp.25-36
• /
• 1995

It is important to separate the compostables from waste for successful operation of composting plant, since various compositions are mixed in it. For the separation of compostables, it is necessary to estimate total amounts of compostables from several sources. Based on it, required capacity and number of composting plant as well as proper waste collection and transportation system can be determined. So, amounts of garbage, major target material for composting. were estimated in this study. In survey of unit garbage generation(UGG). different estimation results would be obtained depending on the basis of its measurement. However, previous researches did not consider it. In this paper, the correlations between area and the number of user of garbage source were analyzed to find the related equations which were apllyed to estimation of total generation. Obtained results are as the following. Relative variations of measured UGG based on area and custumer are 62.5 and 52.8, respectively. In linear regression, related equation between area and custumer was Y=0.244X+59.0 (X=area, Y=custumer). The correlation factor r is 0.904. Equation Y=616.5X/(X+1215.4) was also obtained from linear regression using Monod equation (r=0.720). From the first order equation and measured data of UGG based on custumer, amounts of garbage generation from restaurant in Seoul and the whole country were calculated to 2043.9 ton/d and 9014.0 ton/d, respectively. But, the values calculated from measured data of UGG based on area were as low as 821.3 ton/d Cin Seoul) and 3821.0 ton/d(in the whole country). Consequently, the measurement of unit garbage generation based on the number of custumer was more favorable to lessen the points of survey and to guarantee the representative values. Especially, it would fit well on restaurant having statistics of area.

• Korean Journal of Soil Science and Fertilizer
• /
• v.43 no.3
• /
• pp.329-334
• /
• 2010

The characteristics of nitrogen mineralization in upland soil was studied with 27-week incubation at $25^{\circ}C$. The used soils in this experiment were received six kinds of livestock manure compost each year for four years. Six different composts, which were chicken (CHM), pig (PIM), and cow (COM) manure composted without bulking agent, and chicken (CHMS), pig (PIMS), and cow (COMS) manure composted with sawdust as a bulking agent, were selected for this study. The first-order model was fit to the observed mineral nitrogen (N) vs incubation days using a non-linear regression procedure. The soil potential for N mineralization (No) of manure compost (CHM, PIM, and COM) treated soils were higher than those of the manure-sawdust compost (CHMS, PIMS, and COMS) treated soils. The No value of PIM applied soil was 15.0 mg 100 $g^{-1}$, which was the highest value among the treatments. The amount of N mineralized in compost applied soils ranged from 8.1% to 11.9% of the total N content in soils and increased with increasing total N content in soils. The organic matter content in compost applied soils were negatively correlated with No value (r = $-0.69^*$). Therefore, our result indicated that determination of N application rate in livestock manure compost applied soil should be based on total nitrogen content better than soil organic matter content.

• Journal of Preventive Medicine and Public Health
• /
• v.30 no.1 s.56
• /
• pp.209-228
• /
• 1997

In ordo to investigate the professional job perception and job satisfaction of interns and residents and its related factors, the author conducted survey using self administered questionnaire for 2 months(August and September, 1996). The study subjects were 562 interns and residents who worked at one university hospital(tertiary hospital) and two general hospitals(secondary hospital) in Taegu, Korea. Among them, 297 (52.8%) responded completely. The results were as follows ; Respondents thought that people component was most important, science and status components were next among professional value scales which meant the desirable attitudes required to conduct professional works. But, the score of professional value scales was generally high and not affected by other variables in multiple linear regression analysis. They seemed to have very normative and ideal perceptions about professional values. The score of professionalism scales, which were attitudinal and behavioral traits about professional job, of high grade residents was significantly higher than that of interns and lower grade residents. The score of reference to professional organization and autonomy factor were increasing significantly as the grade increased. Working conditions and perception for socioeconomic status of doctors influenced the score of professionalism scale significantly. It seemed that professional socialization was made during the training periods of interns and residents. Most of respondents answered that current socioeconomic status of doctors were middle and high strata but they responded that the socioeconomic status of doctors would fall in the future. They seemed to have a pessimistic thought about doctor's status. Generally the respondents thought that they were satisfied with doctor job and fit to the job, but 51.9% answered that if possible they would get other jobs. It seemed to reflect their critical thinking on doctors' status. Perceptions about socioeconomic status of doctors, professional value and professionalism influenced job satisfaction significantly. The interns and residents had high sense of calling to doctors and thought that doctors were socially important job. Generally they were satisfied with their job. Interns and residents had normative thoughts about the trait which compentent doctors must have. During the training period, they seemed to have attitude and perception as a professional and to make professional socializations.

• The Korean Journal of Applied Statistics
• /
• v.33 no.3
• /
• pp.335-345
• /
• 2020

This paper deals with the problem of estimating the log-transformed linear regression model to fit actual battle data from the Ardennes Campaign of World War II into the Lanchester model. The problem of determining a global solution for parameters and multicollinearity problems are identified and modified by examining the results of previous studies on data. The least squares method requires attention because a local solution can be found rather than a global solution if considering a specific constraint or a limited candidate group. The method of exploring this multicollinearity problem can be confirmed by a statistic known as a variance inflation factor. Therefore, the Lanchester model is simplified to avoid these problems, and the combat power attrition rate model was proposed which is statistically significant and easy to explain. When fitting the model, the dependence problem between the data has occurred due to autocorrelation. Matters that might be underestimated or overestimated were resolved by the Cochrane-Orcutt method as well as guaranteeing independence and normality.

• Journal of Korean Society of Environmental Engineers
• /
• v.22 no.4
• /
• pp.775-784
• /
• 2000

Nitrate dry deposition fluxes were directly measured using knife-leading-edge surrogate surface (KSS) covered with greased strips and a water surface sampler (WSS). The average gaseous flux ($8.3mg/m^2/day$) was much higher than the average particulate one ($3.0mg/m^2/day$). The best fit gas phase mass transfer coefficient (MTC) of $HNO_3$ was obtained by linear regression analysis between measured gaseous flux containing nitrogen compounds and measured ambient $HNO_3$ concentration. The result showed that the MTCs of $HNO_3$ were approximately two times higher than those of $SO_2$. Especially, during the ozone action day, measured gaseous fluxes containing nitrogen compounds were much higher than those ones calculated as the product of measured ambient $HNO_3$ concentration and gas phase MTC of $HNO_3$, which is calculated from MTC of $SO_2$ using Graham's diffusion law. This result indicated that other nitrogen compounds except $HNO_3$ contributed to gaseous flux containing nitrogen compounds into the water surface sampler. The theoretical calculations suggest the contributions of nitrous acid ($HNO_2$) and PAN to the gaseous dry deposition flux of nitrogen containing compounds to the WSS.

• Korean Journal of Soil Science and Fertilizer
• /
• v.40 no.5
• /
• pp.364-368
• /
• 2007

The objective of this study was to characterize organic mater decomposition with application of livestock manure compost in upland soil. Six different composts, which were chicken (CHM), pig (PIM), and cow (COM) manure compost added and chicken (CHMS), pig (PIMS), cow (COMS) manure compost with sawdust added, were prepared for this study. These composts have different composition of crude carbohydrate (hemicellulose, cellulose, and lignin). The buried-bag method was used to determine the rate of organic mater decomposition and the changes of crude carbohydrate content during 36 months in the field. In all treatment, hemicellulose content was sharply decreased within 8 months, but considerable amount of lignin was remained after 36 months. After 40 months, the rates of carbon decreasing were 81, 80, 72, 69, 67, and 64 % for CHM, PIM, COM, CHMS, PIMS, and COMS, respectively. The estimated equation of carbon decreasing rate (D), $D=aT^b$, was fit to the carbon decreasing rate vs. elapsed time (T) using a non-linear regression procedure. After 40 months, significant difference of carbon decreasing rate between observed and estimated was not found. The relationship between constant a, b and hemicellulose content in the compost was not observed in this experiment. The cellulose and lignin content in the compost were positively correlated to the constant b and negatively correlated to the constant a.