• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.10a
• /
• pp.657-660
• /
• 2005

It is net easy to predict the shrinkage rate of a plastic injection mold in its design process. The shrinkage rate should be considered as one of the important performances to produce the reliable products. The shrinkage rate can be determined by suing the CAE tools in the design produces. However, since the analysis can take minutes to hours, the high computational costs of performing the analysis limit their use in design optimization. Therefore this study was carried out to presume for mutual relation of analysis condition to get the optimum average shrinkage by regression analysis. The results shown that coefficient of determination of regression equation has a fine reliability over 88.3% and regression equation of average shrinkage is made by regression analysis.

• Korean Journal of Veterinary Research
• /
• v.50 no.1
• /
• pp.11-17
• /
• 2010

Linear and non-linear regressions were used to derive the calibration function for the measurement of roxithromycin plasma concentration. Their results were compared with weighted least squares regression by usual weight factors. In this paper the performance of a non-linear calibration equation with the capacity to account empirically for the curvature, y = ax$^{b}$ + c (b $\neq$ 1) is compared with the commonly used linear equation, y = ax + b, as well as the quadratic equation, y = ax$^{2}$+ bx + c. In the calibration curve (range of 0.01 to 10 ${\mu}g/mL$) of roxithromycin, both heteroscedasticity and nonlinearity were present therefore linear least squares regression methods could result in large errors in the determination of roxithromycin concentration. By the non-linear and weighted least squares regression, the accuracy of the analytical method was improved at the lower end of the calibration curve. This study suggests that the non-linear calibration equation should be considered when a curve is required to be fitted to low dose calibration data which exhibit slight curvature.

• Journal of Korean Society on Water Environment
• /
• v.30 no.1
• /
• pp.8-15
• /
• 2014

In this study, regression equation was analyzed to estimate non-point source (NPS) pollutant loads in orchard area. Many factors affecting the runoff of NPS pollutant as precipitation, storm duration time, antecedent dry weather period, total runoff density, average storm intensity and average runoff intensity were used as independent variables, NPS pollutant was used as a dependent variable to estimate multiple regression equation. Based on the real measurement data from 2008 to 2012, we performed correlation analysis among the environmental variables related to the rainfall NPS pollutant runoff. Significance test was confirmed that T-P ($R^2=0.89$) and BOD ($R^2=0.79$) showed the highest similarity with the estimated regression equations according to the NPS pollutant followed by SS and T-N with good similarity ($R^2$ >0.5). In the case of regression equation to estimate the NPS pollutant loads, regression equations of multiplied independent variables by exponential function and the logarithmic function model represented optimum with the experimented value.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.52 no.6
• /
• pp.1-7
• /
• 2010

This study was carried out to investigate the effect of hydro power factors (e.g., irrigation area, watershed area, active storage, gross head) on annual generation capacity and operation ratio for agricultural reservoirs in Chungbuk Province with active storage of over 1 million $m^3$. The annual generation capacity and operation ratio were estimated using HOMWRS (Hydrological Operation Model for Water Resources System) from last 10-year daily hydrological data. The correlation coefficients between annual generation capacity and the hydro power factors except gross head were high (over 0.87), but the correlation coefficients between operational rate and the factors were low (below 0.28). The optimum multiple regression equations of the annual generation capacity were expressed as the functions of watershed area, active storage, and gross head. Also, the simple regression equation of annual generation capacity was expressed as a function of watershed area. The average relative root-mean-square-error (RRMSE) between observed and estimated values by the optimum multiple regression equations was smaller than that by the simple regression equation, suggesting that the former has more accuracy than the latter.

• 모자간호학회지
• /
• v.1
• /
• pp.34-44
• /
• 1991

The purposes of this study were to predict newborn birthweight by use of gestational period and fundal height and to identify growth curve of fundal height according to gestational period and growth curve of newborn birthweight according to fundal height. The subjects for the study were 802 women who delivered the normal newborn babies at Seoul National University Hospital from Sep. 1, 1981 to Aug.31, 1986. The data were collected bit chart review and analyzed nth SPSS program. The results of study were as follows : 1. The multiple regression equation ($R^2$=0.416) used for the prediction of newborn birthweight was y=(newborn birthweight, kg)=-4.421+0.075$x_1$(fundal height, cm)+0.053$x_2$(gestational period, weeks)+0.016$x_3$(abdominal girth, cm)+0.010$x_4$(maternal height, cm) 2. The growth curve of fundal height according to gestational period was obtained by polynomial regression. The regression equation was Y(fundal height, cm)=-36.78+18.58$log_ex$(gestational period, weeks) The growth curve of newborn birth weight according to fundal height was obtained by polynomial regression. The regression equation was Y(newborn birthweight, kg)=-8.09+3.27$log_ex$ (Fundal Height, cm) 3. In the following subgroups no significant difference was found in fundal height : engaged vs. nonengaged presentation, and nulliparous vs. multiparous women.

• Journal of Biomedical Engineering Research
• /
• v.30 no.5
• /
• pp.401-411
• /
• 2009

In this paper, we correct pulse wave velocity(PWV) with heart-rate and derive regression equations to estimate intima-media thickness(IMT). Widely used methods for diagnosis of arteriosclerosis are IMT and PWV. Arterial wall stiffness determines the degree of energy absorbed by the elastic aorta and its recoil in diastole but there is not correlation between sclerosis and IMT in an existing study. In this study, we will correct PWV with heart-rate and get regression equation to estimate IMT using heart-rate correction index(HCI). We executed experiments for this study. Made up question of physical condition and measured electrocardiogram(ECG), photoplethysmogram (PPG) of finger-tip and toe-tip and ultrasound image of carotid artery. Calculated PWV and IMT using ECG, PPG and ultrasound image. We found that every p-value between PWV and IMT is not significant(<0.05). But p-value between IMT and HCI which is a corrected PWV using heart-rate is significant(>0.01). We use HCI and various measured parameter for estimating regression equation and apply backward estimation to select parameters for regression analysis. Result of backward estimation, found that only HCI is possible to derive proper regression equation of IMT. Relationship between PWV and IMT is the second order. Result of regression equation of E-H PWV is $R^2$=0.735, adj $R^2$=0.711. This is the best correlation value. We calculate error of its analysis for verification of earlobe PWV regression equation. Its result is RMSEP=0.0328, MAPE(%) = 4.7622. Like this regression analysis, we know that HCI is useful parameter and relationship between PWV, HCI and IMT. In addition, we are able to suggest possibility which is that we can get different parameter of prediction throughout just one measurement.

• Journal of Korea Water Resources Association
• /
• v.55 no.4
• /
• pp.257-266
• /
• 2022

Inundation damage is increasing every year due to localized heavy rain and an increase of rainfall exceeding the design frequency. Accordingly, the importance of hydraulic structures for flood control and defense is also increasing. The hydraulic structures are designed according to its purpose and performance, and the amount of flood is an important calculation factor. However, in Korea, design rainfall is used as input data for hydrological analysis for the design of hydraulic structures due to the lack of sufficient data and the lack of reliability of observation data. Accurate probability rainfall and its temporal distribution are important factors to estimate the design rainfall. In practice, the regression equation of temporal distribution for the design rainfall is calculated using the cumulative rainfall percentage of Huff's quartile method. In addition, the 6^th order polynomial regression equation which shows high overall accuracy, is uniformly used. In this study, the optimized regression equation of temporal distribution is derived using the variable selection method according to the principle of parsimony in statistical modeling. The derived regression equation of temporal distribution is verified through the significance test. As a result of this study, it is most appropriate to derive the regression equation of temporal distribution using the stepwise selection method, which has the advantages of both forward selection and backward elimination.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2001.10a
• /
• pp.509-512
• /
• 2001

In this study, pollutant load from paddy field was estimated by regression equation from 5 to 8 in 2001. During study period, total rainfall was 511.3mm and runoff discharge was 968.71mm. Regression equation between flow rate(m3/s) and pollutant loading rate(g/s) is exponential relationship. For site 1, coefficient of determination (R2) for $COD_{cr}$, T-P, T-N were 0.7068, 0.8441, 0.6806 respectively and site 2, 0.9369, 0.8855, 0.4262 respectively. Considering unit loads, Jun was the highest valus as 13.85 $COD_{c}kg/km2/day$, 0.24 T-Pkg/km2/day, 1.22 T-Nkg/km2/day. Until study period, total $COD_{cr}$ load estimated regression equation is 19.32kg/km2/day and, T-P, T-N were 0.264, 1.88 respectively

• Journal of Sericultural and Entomological Science
• /
• v.20 no.2
• /
• pp.36-39
• /
• 1978

With the current advances in insect toxicant bioassay, the need for easy methods of estimating the dosage-mortality regression equation has become vital. The Probit analysis seems to be not convenient for estimating the dosage-mortality regression equation and median lethal dose(LD50) because of its complexity in calculation. This study presents a comparision between Probit and Losit transformation for the estimation from bioassay results. Validation of the two methods is presented for the pathogenecity of nuclear polyhedrosis virus to the larva of fall web worm, Hyphantria cunea D.

• Journal of Preventive Medicine and Public Health
• /
• v.28 no.1 s.49
• /
• pp.73-84
• /
• 1995

This study was conducted to measure the lead level in the blood, scalp hair and toenail of the elementary schoolchildren and assess the relationship among those samples. Lead concentration of the blood, scalp hair and toenail was measured for 100(male 50, female 50) fourth grade elementary schoolchildren in Taegu city. The mean lead level in the blood, scalp hair and toenail was $6.00{\pm}2.44{\mu}g/dl,\;6.68{\pm}3.54{\mu}g/g,\;and\;7.33{\pm}3.18{\mu}g/g. The mean lead level in the blood of schoolboys was $6.43{\pm}2.77{\mu}g/dl$, and that of schoolgirls was $5.59{\pm}2.01{\mu}g/dl$. The mean lead level in the scalp hair of schoolboys was $7.66{\pm}2.97{\mu}g/g$ and that of schoolgirls was $6.88{\pm}3.54{\mu}g/g$. The mean lead level in the toenail of schoolboys was $8.19{\pm}3.5{\mu}g/g$ and that of schoolgirls was $6.47{\pm}2.52{\mu}g/g$ and their difference was statistically significant. In schoolboys, the correlation coefficient between the lead level in the blood and scalp hair was 0.4909, and the data were fitted best by the regression equation Y = 0.5255X+4. 2810, where Y and X are scalp hair and blood concentration. In schoolgirls the correlation coefficient between the lead level in the blood and scalp hair was 0.3778, and the data were fitted best by the regression equation Y = 0.6655X+2.9632, where Y and X are scalp hair and blood concentration. In schoolboys, the correlation coefficient between the lead level in the blood and in the toenail was 0.5533, and the data were fitted best by the regression equation Y = 0.7076X+3. 6472, where Y and X are toenail and blood concentration. In schoolgirls the correlation coefficient between the lead level in the blood and in the toenail was 0.2738, and the data were fitted best by the regression equation Y = 0.3431X+4.5570, where Y and X are toenail and blood concentration In schoolboys, the correlation coefficient between the lead level in the scalp hair and in the toenail, in the schoolboys was 0.4148, and the data were fitted best by the regression equation Y = 0.4956X+4.3986, where Y and X are toenail and scalp hair concentration. In schoolgirls, the correlation coefficient between the lead level in the scalp hair and in the toenail was 0.1159, and the data were fitted best by the regression equation Y = 0.0825X+5. 9214, where Y and X are toenail and scalp hair concentration. Correlation among lead concentration in the blood, scalp hair and toenail of schoolchildren were statistically significant except between scalp hair and toenail in schoolgirls. These finding suggest that blood, scalp hair and toenail can be used as substitutive samples between each others.