• The Korean Journal of Applied Statistics
• /
• v.33 no.6
• /
• pp.805-812
• /
• 2020

High-frequency data are now prevalent in financial time series. As a functional data arising from high-frequency financial time series, we are concerned with the intraday volatility to which functional principal component analysis (FPCA) is applied in order to achieve a dimension reduction. A review on FPCA and R function is made and high-frequency KOSPI volatility is analysed as an application.

• Journal of the Korean Mathematical Society
• /
• v.54 no.1
• /
• pp.87-99
• /
• 2017

To indicate the statistical complexity of dynamical systems, we introduce the notions of higher order irregular set and higher order maximal Birkhoff average oscillation in this paper. We prove that, in the setting of topologically mixing Markov chain, the set consisting of those points having maximal k-order Birkhoff average oscillation for all positive integers k is as large as the whole space from the topological point of view. As applications, we discuss the corresponding results on a repeller.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.539-554
• /
• 2018

Effective interaction between user and device is considered an important ability of IoT devices. For some applications, it is necessary to recognize human facial expressions in real time and make accurate judgments in order to respond to situations correctly. Therefore, many researches on facial image analysis have been preceded in order to construct a more accurate and faster recognition system. In this study, we constructed an automatic recognition system for facial expressions through two steps - a facial recognition step and a classification step. We compared various models with different sets of data with pixel information, landmark coordinates, Euclidean distances among landmark points, and arctangent angles. We found a fast and efficient prediction model with only 30 principal components of face landmark information. We applied several prediction models, that included linear discriminant analysis (LDA), random forests, support vector machine (SVM), and bagging; consequently, an SVM model gives the best result. The LDA model gives the second best prediction accuracy but it can fit and predict data faster than SVM and other methods. Finally, we compared our method to Microsoft Azure Emotion API and Convolution Neural Network (CNN). Our method gives a very competitive result.

• Journal of Communications and Networks
• /
• v.14 no.1
• /
• pp.1-9
• /
• 2012

We consider a receiving node, located at the origin, and a Poisson point process (PPP) that models the locations of the desired transmitter as well as the interferers. Interference is known to be non-Gaussian in this scenario. The capacity bounds for additive non-Gaussian channels depend not only on the power of interference (i.e., up to second order statistics) but also on its entropy power which is influenced by higher order statistics as well. Therefore, a complete statistical characterization of interference is required to obtain the capacity bounds. While the statistics of sum of signal and interference is known in closed form, the statistics of interference highly depends on the location of the desired transmitter. In this paper, we show that there is a tradeoff between entropy power of interference on the one hand and signal and interference power on the other hand which have conflicting effects on the channel capacity. We obtain closed form results for the cumulants of the interference, when the desired transmitter node is an arbitrary neighbor of the receiver. We show that to find the cumulants, joint statistics of distances in the PPP will be required which we obtain in closed form. Using the cumulants, we approximate the interference entropy power and obtain bounds on the capacity of the channel between an arbitrary transmitter and the receiver. Our results provide insight and shed light on the capacity of links in a Poisson network. In particular, we show that, in a Poisson network, the closest hop is not necessarily the highest capacity link.

• Women's Health Nursing
• /
• v.27 no.1
• /
• pp.40-48
• /
• 2021

Purpose: This study compared maternal age at childbirth, the number of live births, and the weight of live births between Korean women and immigrant women using statistical data from the Republic of Korea for the period of 2008-2018. Methods: The analysis was conducted using data from the Microdata Integrated Service of Statistics Korea (https://mdis.kostat.go.kr/index.do). Results: Korean women and immigrant women showed a higher age at childbirth in 2018 than in 2008. The percentage of newborns of Korean women with a birth weight of less than 2.5 kg increased slightly for 3 consecutive years from 2016 to 2018, whereas for immigrant women, this percentage increased in 2017 compared to 2016 and then decreased again in 2018. Very low birth weight (less than 1.5 kg) became more common among immigrant women from 2016 to 2018. Birth at a gestational age of fewer than 37 weeks increased both among Korean and immigrant women from 2016 to 2018. In both groups, the percentage of women who had their first child within their first 2 years of marriage decreased from 2008 to 2018. Conclusion: Immigrant women had higher birth rates than Korean women, while both groups showed an increasing trend in preterm birth. Greater attention should be paid to the pregnancy and birth needs of immigrant women, and steps are needed to ensure health equity and access in order to prevent preterm births. It is also necessary to identify factors that affect preterm birth and birth of very low birth weight infants among immigrant women in the future.

• Journal of applied mathematics & informatics
• /
• v.13 no.1_2
• /
• pp.457-470
• /
• 2003

The comparison of two or more Lorenz curves of Pareto distributions of first kind under arbitrary order restriction is studied. The problem is turned out to be a statistical inference problem concerning scale parameters under order restriction. We assume that the location parameters of Palate distributions are completely unknown. In this paper the maximum likelihood estimation and likelihood ratio tests for and against order restriction are proposed.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.4
• /
• pp.981-992
• /
• 2004

In this paper, we develop the noninformative priors for the common scale parameter in the inverse gaussian distributions. We developed the first and second order matching priors. Next we revealed that the second order matching prior satisfies a HPD matching criterion. Also we showed that the second order matching prior matches alternative coverage probabilities up to the second order. It turns out that the one-at-a-time reference prior satisfies a second order matching criterion. Some simulation study is performed.

• Kyungpook Mathematical Journal
• /
• v.60 no.4
• /
• pp.797-803
• /
• 2020

The Brück conjecture is still open for an entire function f with hyper order of no less than 1/2, which is not an integer. In this paper, it is proved that the hyper order of solutions of a linear complex differential equation that is related to the Brüuck Conjecture is infinite. The results show that the conjecture holds in a special case when the hyper order of f is 1/2.

• Journal of the Korean Statistical Society
• /
• v.32 no.2
• /
• pp.193-202
• /
• 2003

We consider a double threshold AR-ARCH type process and give sufficient conditions under which the higher-order moments exist. Geometric ergodicity and strict stationarity are also studied.

• Honam Mathematical Journal
• /
• v.26 no.4
• /
• pp.365-378
• /
• 2004

The orders in quaternion algebras play central role in the theory of Hecke operators. In this paper, we study the order of two sided ideal group in orders of a quaternion algebra.