• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.177-183
• /
• 1997

The cumulative function of the noncentral t distribution calculate power in testing equality of means of two normal populations and confidence intervals for the ratio of population mean to standard deviation. In this paper, the evaluation of the cumulative function of noncentral t distribution is applied to the neural network consists of the multi-layer perception structure and learning process has the algorithm of the backpropagation. Numerical comparisons are made between the Fisher's values and the results obtained by neural network theory.

• Communications for Statistical Applications and Methods
• /
• v.19 no.2
• /
• pp.277-286
• /
• 2012

For credit assessment models, the ROC curves evaluate the classification performance using two univariate cumulative distribution functions of the false positive rate and true positive rate. In this paper, it is extended to two bivariate normal distribution functions of default and non-default borrowers; in addition, the bivariate ROC curves are proposed to represent the joint cumulative distribution functions by making use of the linear function that passes though the mean vectors of two score random variables. We explore the classification performance based on these ROC curves obtained from various bivariate normal distributions, and analyze with the corresponding AUROC. The optimal threshold could be derived from the bivariate ROC curve using many well known classification criteria and it is possible to establish an optimal cut-off criteria of bivariate mixture distribution functions.

• Wind and Structures
• /
• v.36 no.4
• /
• pp.249-261
• /
• 2023

Aerodynamic force coefficients are generally prescribed by an ensemble average of ten and/or twenty 10-minute samples. However, this makes it difficult to identify the exact probability distribution and exceedance probability of the prescribed values. In this study, 12,600 10-minute samples on three tall buildings were measured, and the probability distributions were first identified and the aerodynamic force coefficients corresponding to the specific non-exceedance probabilities (cumulative probabilities) of wind load were then evaluated. It was found that the probability distributions of the mean and fluctuating aerodynamic force coefficients followed a normal distribution. The ratios of aerodynamic force coefficients corresponding to the specific non-exceedance probabilities (C_f,Non) to the ensemble average of 12,600 samples (C_f,Ens), which was defined as an adjusting factor (C_f,Non/C_f,Ens), were less than 2%. The effect of coefficient of variation of wind speed on the adjusting factor is larger than that of the annual non-exceedance probability of wind load. The non-exceedance probabilities of the aerodynamic force coefficient is between P_C,nonex = 50% and 60% regardless of force components and aspect ratios. The adjusting factors from the Gumbel distribution were larger than those from the normal distribution.

• Journal of ILASS-Korea
• /
• v.20 no.3
• /
• pp.175-180
• /
• 2015

A series of sprinkler discharging tests was conducted to measure the droplet size and its distribution of residential fire sprinkler heads. Droplet sizes in sprinkler spray were measured using a laser diffraction method for the flush, circular and pendent type sprinkler head. In this study, the $D_{v,50}$ of the flush type sprinkler heads were ranged between $530{\sim}1040{\mu}m$ and those of circular and pendent type were $988{\mu}m$ and $916{\mu}m$, respectively. The measured cumulative volume distributions were followed by a combination of the log-normal and Rosin-Rammler distribution which is widely used in the computational fire analysis and the parameters of distribution function were obtained from the best fit line through the measured data.

• Journal of Preventive Medicine and Public Health
• /
• v.17 no.1
• /
• pp.95-106
• /
• 1984

To acquire the essential basic data to the establishment of control measure for the hazardous health effect that could be caused by harmful metals, the author measured the concentrations of trace metals in whole blood of women of $20{\sim}39$ years old living in urban and rural area using atomic absorption spectrophotometer. The summarized results were as follows; 1. The mean concentration of zinc in whole blood was $10.69{\pm}8.07{\mu}g/ml$ in rural area. The frequency distribution by zinc concentration level was nearly L-type and the cumulative frequency distribution was showed bimodal type in both area. 2. The mean iron concentration in whole blood was $323.09{\pm}87.15{\mu}g/ml$ and $322.07{\pm}104.74{\mu}g/ml$ in urban and rural area, respectively. The frequency distribution was similar to normal distribution type in both area, but the cumulative distribution was unimodal type in urban area and bimodal type in rural area. 3. The mean magnesium concentration was $41.08{\pm}19.58{\mu}g/ml$ and $40.28{\pm}16.82{\mu}g/ml$ in the area, respectively. The frequency distribution type had skewness to the right and the cumulative frequency distribution was unimodal type in both area. 4. The mean copper concentration was $1.417{\pm}0.761{\mu}g/ml$ and $1.375{\pm}0.743{\mu}g/ml$ in the area, respectively. The frequency distribution type had skewness to the right and the cumulative frequency distribution was bimodal type in both area. 5. The mean manganese concentration was $0.079{\pm}0.039{\mu}g/ml$ and $0.07{\pm}0.058{\mu}g/ml$ in the area, respectively. The frequency distribution type had skewness to the right in both area but slight irregular in rural area and the cumulative distribution was unimodal and bimodal type in urban and rural area, respectively. 6. The mean cadmium concentration in whole blood was $0.031{\pm}0.026{\mu}g/ml$ in urban and $0.028{\pm}0.023{\mu}g/ml$ in rural area. The frequency distribution type had skewness to the right and cumulative frequency distribution was bimodal type in both area.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.33 no.12
• /
• pp.1482-1487
• /
• 2009

In this study statistical analysis was performed for NDI(Non-Destructive Inspection) results of F100 engine front seal support assembly. NDI results can be statistically considered as Quantal Response Data. It is found that the suitable probability distribution to the failure data is normal distribution through MLE(Maximum Likelihood Estimation) of the Quantal Response Data. Moreover, Cumulative Distribution Function, failure rate function and B-Life are calculated on the supposed distribution.

• Journal of the Korean Statistical Society
• /
• v.33 no.1
• /
• pp.107-128
• /
• 2004

Bivariate Laplace distributions for which both marginal distributions and Laplace are discussed. Three kinds of bivariate Laplace distributions which are extended bivariate exponential distributions of Gumbel (1960) are introduced in this paper. These symmetrical distributions are compared with asymmetrical distributions of Kotz et al. (2000). Their probability density functions, cumulative distribution functions are derived. Conditional skewnesses and kurtoses are also defined. Their correlation coefficients are calculated and compared with others. We proposed bivariate random vector generating methods whose distributions are bivariate Laplace. With sample means and medians obtained from generated random vectors, variance and covariance matrices of means and medians are calculated and discussed with those of bivariate normal distribution.

• Geotechnical Engineering
• /
• v.1 no.2
• /
• pp.41-54
• /
• 1985

Generally, ground settlements and lateral displacements are accompanied by underground excavation associated with open-cut or tunnling. These ground movements cause a harmful influence upon nearby super.structures and sub-structures. Occasionally, the ground movements may pose serious problems as the function of the nearby structures may be disrupted. Therefore, prior to the subway construction in an urban area, it is necessary to identify the causes of ground settlements and estimating the extent St the magnitude of ground movements since any potential damage to the nearby structures such as gas lines, water mains, high buildings and cultural assets must be assessed. The research was performed mainly on ground movements such as surface settlements, lateral displacements, subsurface settlements and crown settlements to predict the maximum settlement and settlement zone, and to identify the causes of ground settlements in NATM sections of Busan subway. As a result, it was found that lateral distribution of settlements could be approximated reasonably by a Gaussian normal probability curve and longitudinal distribution of settlements by a cumulative Gaussian probability curve, and that the early closure of temporary invert was very important to minimize ground settlements.

• YAKHAK HOEJI
• /
• v.35 no.5
• /
• pp.379-383
• /
• 1991

This study was attempted to investgate the pharmacokinetics of theophylline (4 mg/kg i.v) in the rabbits pretreated with propranolol (1 and 2.5 mg/kg/hr, infusion) for four hours. The plasma concentration and AUC of theophylline were increased in rabbits pretreated with propranolol as compared with those of normal rabbits. The amount of cumulative urinary excretion and renal clearance and total body clearance were decreased in rabbits pretreated with propranolol as compared with those of normal rabbits. The apparent volume of distribution was slightly affected by change of the clearance of theophylline. From the results of this experiment, it is desirable that dosage regimen of theophylline should be adjusted when theophylline combined with propranolol in clinical pharmacy practice.

• Journal of Korean Society for Quality Management
• /
• v.23 no.1
• /
• pp.15-28
• /
• 1995

We consider a step-stress life testing which is devised for a two-component parallel system with considerably long life time. To describe such a system, we use an exponential distribution as the survival function. The lift distribution is assumed between the log mean life time and the stress with the cumulative exposure model. The criterion for optimality is to minimize the sum of the variances of the maximum likelihood estimators of the mean life times of each part under the normal stress.