• Communications for Statistical Applications and Methods
• /
• v.29 no.3
• /
• pp.333-351
• /
• 2022

The skew-t distribution is an attractive family of asymmetrical heavy-tailed densities that includes the normal, skew-normal and Student's-t distributions as special cases. In this work, we propose an EM-type algorithm for computing the maximum likelihood estimates for skew-t linear regression models with censored response. In contrast with previous proposals, this algorithm uses analytical expressions at the E-step, as opposed to Monte Carlo simulations. These expressions rely on formulas for the mean and variance of a truncated skew-t distribution, and can be computed using the R library MomTrunc. The standard errors, the prediction of unobserved values of the response and the log-likelihood function are obtained as a by-product. The proposed methodology is illustrated through the analyses of simulated and a real data application on Letter-Name Fluency test in Peruvian students.

• Nuclear Engineering and Technology
• /
• v.44 no.7
• /
• pp.755-766
• /
• 2012

A Binary Decision Diagram (BDD) is a graph-based data structure that calculates an exact top event probability (TEP). It has been a very difficult task to develop an efficient BDD algorithm that can solve a large problem since its memory consumption is very high. Recently, in order to solve a large reliability problem within limited computational resources, Jung presented an efficient method to maintain a small BDD size by a BDD truncation during a BDD calculation. In this paper, it is first identified that Jung's BDD truncation algorithm can be improved for a more practical use. Then, a more efficient truncation algorithm is proposed in this paper, which can generate truncated BDD with smaller size and approximate TEP with smaller truncation error. Empirical results showed this new algorithm uses slightly less running time and slightly more storage usage than Jung's algorithm. It was also found, that designing a truncation algorithm with ideal features for every possible fault tree is very difficult, if not impossible. The so-called ideal features of this paper would be that with the decrease of truncation limits, the size of truncated BDD converges to the size of exact BDD, but should never be larger than exact BDD.

• BMB Reports
• /
• v.41 no.12
• /
• pp.881-885
• /
• 2008

Protein phosphatase 1 consists of a secondary structure arrangement, conserved in the serine/threonine protein phosphatase gene family, flanked by nonconserved N-terminal and C-terminal domains. The deletion mutant of PP1 with the 8 nonconserved N-terminal residues removed was designated PP1-(9-330). PP1-(9-330) had a higher activity and affinity than PP1 when assayed against four different substrates, and it also demonstrated a 6-fold higher sensitivity to the inhibitor okadaic acid. This suggested that the N-terminal domain suppresed the activity of PP1 and interfered with its inhibition by okadaic acid. The ANS fluorescence intensity of PP1-(9-330) was greater than that of PP1, which implies that the hydrophobic groove running from active site in the truncated PP1 was more hydrophobic than in PP1. Our findings provide evidence that the nonconserved N-terminus of PP1 functions as an important regulatory domain that influences the active site and its pertinent properties.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.6
• /
• pp.495-507
• /
• 2009

The expected sliding distance for the lifetime of a caisson breakwater has a limitation to be used as the stability criterion of the breakwater. Since the expected sliding distance is calculated as the mean of simulated sliding distances for the lifetime, there is possibility for the actual sliding distance to exceed the expected sliding distance. To overcome this problem, the exceedance probability of the allowable sliding distance is used to assess the stability of sliding. Latin Hypercube sampling and Crude Monte Carlo simulation were used to calculate the exceedance probability. The doubly-truncated normal distribution was considered to complement the physical disadvantage of the normal distribution as the random variable distribution. In the case of using the normal distribution, the cross-sections of Okgye, Hwasun, and Donghae NI before reinforcement were found to be unstable in all the limit states. On the other hand, when applying the doubly-truncated normal distribution, the cross-sections of Hwasun and Donghae NI before reinforcement were evaluated to be unstable in the repairable limit state and all the limit states, respectively. Finally, the shortcoming of the expected sliding distance as the stability criterion was investigated, and we reasonably assessed the stability of sliding of caissons by using the exceedance probability of allowable sliding distance for the caisson breakwaters in Korea.

• Korean Management Science Review
• /
• v.24 no.1
• /
• pp.77-90
• /
• 2007

Taguchi assumed that a product characteristic has the uniform distribution in its preventive maintenance limit when deriving the expected loss generated by the quality deviation. But it is reasonable to assume that a product characteristic has the normal distribution than the uniform distribution. On this paper, we first find the optimum inspection interval and the optimum preventive maintenance limit under the truncated triangular distribution. Secondly we use the beta-general distribution and compare with the truncated triangular distribution. By using the numerical examples, we find the optimum inspection interval and the optimum preventive maintenance limit under their distributions. As a result, we find that the beta-general distribution gives the best solution and easy calculation.

• Tuberculosis and Respiratory Diseases
• /
• v.46 no.3
• /
• pp.327-337
• /
• 1999

Background: Reactive oxygen species are involved in multi-stage process of carcinogenesis. The moot of cancer cell lines and cancer cells in tumor tissue produce reactive oxygen species and on the other hand, the activities of catalase, Mn- and CuZn-superoxide dismutase in tumor cells are usually low. These persistent oxidative stress in tumor tissue facilitates tumor invasion and metastasis. 12-kDa thioredoxin, which regulates the intracellular redox potential with glutathione and glutaredoxin is involved in cell activation, proliferation, differentiation and redox-mediated apoptosis. It is also purified as 14-kDa and 10-kDa eooinophilic cytotoxic enhancing factor(ECEF) from human histiocytic cell(U937) and 10-kDa ECEF has more than 20 times eosinophilic stimulation activity than 14-kDa ECEF. It has been reported that adult T-cell leukemia, squamous cell carcinoma of uterine cervix, and hepatocellular carcinoma show increased amounts of human thioredoxin and thioredoxin mRNA is increased in lung cancer. In this study, we investigated the expression of conventional antioxidant enzymes such as catalase, CuZn-SOD, and glutathione peroxidase and thioredoxin in lung cancer tissue compared to adjacent normal lung tissue and the induction of thioredoxin in macrophage cells after treatment of oxidative stress and endotoxin Methods: We measured the amount of conventional antioxidant enzymes such as catalase, CuZn-SOD, and glutathione peroxidase and thioredoxin in lung cancer tissue compared to adjacent normal lung tissue by immunoblot analysis and the induction of thioredoxin in mouse monocyte-macrophage cells(RAW 264.7) by treatment of 5 ${\mu}M$ menadione and 1 ${\mu}g/ml$ endotoxin Results: On immunoblot analysis, the expression of 12-kDa thioredoxin was increased in lung cancer tissue compared to paired normal lung tissue. but the expression of catalase and CuZn-SOD were decreased in lung cancer tissue compared to paired normal tissue and the expression of glutathione peroxidase in lung cancer was variable. The expression of truncated thioredoxin was also increased in lung cancer. When mouse monocyte-macrophage cells were treated with 5 ${\mu}M$ menadione and 1 ${\mu}g/ml$ endotoxin, the expression of thioredoxin was peaked at 12 hrs and sustained to 48 hrs. Conclusion: In contrast with other conventional antioxidants, the expression of 12-kDa and truncated thioredoxin in lung cancer were increased and it is closely associated with persistent oxidative stress in tumor microenvironment. Considering especially the biological functions of truncated thioredoxin, the increased amount of truncated thioredoxin has significant role in tumor growth through cell proliferation.

• Design Convergence Study
• /
• v.16 no.3
• /
• pp.17-37
• /
• 2017

The purpose of this study is to evaluate the value of the center of age-friendly goods to confirm the public value of the center. 152 elderly participants, either who have utilized the center themselves or who have experienced the center indirectly through the image have participated in the study. The experience of age-friendly product center was evaluated by the contingent valuation method, which is widely used in the public services. A double-bounded dichotomous choice (DB-DC) log-logistic model and a log-normal model were conducted. As the result of the estimation, the truncated mean of the log-logistic model was 3,401 Korean won, with the mean of 4,937 won. In addition, the truncated mean of the log-normal model was 3,433 won, with the mean of 4,144 won. Elderly who have not experienced the center placed a higher value on the center, compared to the experienced elderly. The results imply the necessity to enlarge the opportunity for future suppliers to experience the center for age-friendly products, in order to improve their objective understandings. In addition, the necessity of expanding exhibition products and applying age-friendly designs is proposed, to offer various attractive experiences to elderly.

• Journal of Applied Reliability
• /
• v.4 no.1
• /
• pp.15-26
• /
• 2004

Life test is performed to set a confidence (lower) limit on the mean or median life of items if the number of failures at the end of the fixed time t does not exceed a given number c. Gupta(1962) propose a sampling plan for truncated life tests when the life distribution of an item is normal or lognormal distribution. In this paper, based on the result of Gupta(1962), we propose a sampling plan for failure rate test when an item has normal or lognormal life distribution. We assume that the shape parameter is known while the location parameter is unknown.

• Industrial Engineering and Management Systems
• /
• v.11 no.4
• /
• pp.390-396
• /
• 2012

The present paper extends the idea of tightened-normal-tightened sampling scheme to group acceptance sampling plans under the time truncated life tests. We consider three famous distributions that are widely used in the area of reliability such as the generalized exponential distribution, the Weibull distribution, and the Birnbaum-Saunders distribution in the proposed sampling plan. The plan parameters are determined such that the producer's risk and the consumer's risk are satisfied at the specified median life. Extensive tables showing plan parameters are provided at various values of the experiment time and the consumer's risk for each of three distributions for the practical use. Some examples are given to illustrate the procedure of the proposed plan.

• The Korean Journal of Applied Statistics
• /
• v.32 no.5
• /
• pp.693-702
• /
• 2019

Deep learning has gained popularity for the classification and prediction task. Neural network layers become deeper as more data becomes available. Saturation is the phenomenon that the gradient of an activation function gets closer to 0 and can happen when the value of weight is too big. Increased importance has been placed on the issue of saturation which limits the ability of weight to learn. To resolve this problem, Glorot and Bengio (Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 249-256, 2010) claimed that efficient neural network training is possible when data flows variously between layers. They argued that variance over the output of each layer and variance over input of each layer are equal. They proposed a method of initialization that the variance of the output of each layer and the variance of the input should be the same. In this paper, we propose a new method of establishing initialization by adopting truncated normal distribution and truncated cauchy distribution. We decide where to truncate the distribution while adapting the initialization method by Glorot and Bengio (2010). Variances are made over output and input equal that are then accomplished by setting variances equal to the variance of truncated distribution. It manipulates the distribution so that the initial values of weights would not grow so large and with values that simultaneously get close to zero. To compare the performance of our proposed method with existing methods, we conducted experiments on MNIST and CIFAR-10 data using DNN and CNN. Our proposed method outperformed existing methods in terms of accuracy.