• Communications for Statistical Applications and Methods
• /
• v.26 no.5
• /
• pp.497-506
• /
• 2019

Forecasting the U.S. employment level is made using machine learning methods of the artificial neural network: deep neural network, long short term memory (LSTM), gated recurrent unit (GRU). We consider the big data of the federal reserve economic data among which 105 important macroeconomic variables chosen by McCracken and Ng (Journal of Business and Economic Statistics, 34, 574-589, 2016) are considered as predictors. We investigate the influence of the two statistical issues of the dimension reduction and time series differencing on the machine learning forecast. An out-of-sample forecast comparison shows that (LSTM, GRU) with differencing performs better than the autoregressive model and the dimension reduction improves long-term forecasts and some short-term forecasts.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.3
• /
• pp.541-550
• /
• 2015

The importance of efficient enterprise network management has been emphasized continuously because of the rapid utilization of Internet in a limited resource environment. For the efficient network management, the management policy that reflects the characteristics of a specific network extracted from long-term traffic analysis is essential. However, the long-term traffic data could not be handled in the past and there was only simple analysis with the shot-term traffic data. However, as the big data analytics platforms are developed, the long-term traffic data can be analyzed easily. Recently, enterprise network resource efficiency through the long-term traffic analysis is required. In this paper, we propose the methods of collecting, storing and managing the long-term enterprise traffic data. We define several classification categories, and propose a novel network resource efficiency through the multidirectional statistical analysis of classified long-term traffic. The proposed method adopted to the campus network for the evaluation. The analysis results shows that, for the efficient enterprise network management, the QoS policy must be adopted in different rules that is tuned by time, space, and the purpose.

• Health Policy and Management
• /
• v.30 no.4
• /
• pp.522-530
• /
• 2020

Background: In 2008, Korea implemented a new type of social insurance known as "long-term care insurance". We examined the association between ownership of long-term care facilities and the incidence of pressure ulcers after the implementation of "long-term care insurance". This study is a population-based retrospective cohort study from 2006 to 2013. Methods: We used medical claims data from the Korean National Health Insurance Corporate Elderly Cohort Database from 2006 to 2013. These data comprise a nationally representative sample. To avoid confounders, only patients admitted to one long-term care facility and who stayed for >70% of the follow-up time were included; as a result, 3,107 individuals were enrolled. The main independent variable was the operating entity of the long-term care facility (local government, corporate bodies, and private for-profit owners), and the dependent variable was the 1-year incidence of pressure-ulcers. Survival analysis (Cox proportional hazard model) was used as an analysis method. Results: Compared to patients admitted to local government long-term care facilities, patients admitted to private long-term care facilities had a significantly higher 1-year risk of pressure ulcers (hazard ratio [HR], 1.94; 95% confidence interval [CI], 1.29-2.91); the risk was especially high among patients who were cognitively dependent (HR, 2.34; 95% CI, 1.25-4.37). Conclusion: Patients admitted to private for-profit long-term care facilities were more likely to have pressure ulcers compared to those in local government and corporate body long-term care facilities. Appropriate assessment tools and publicly available information, as well as more restricted legal requirements, are needed to improve the care quality and outcomes of patients in long-term care facilities.

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.2
• /
• pp.421-429
• /
• 2008

We consider a general semiparametric additive risk model that consists of three components. They are parametric, purely and smoothly nonparametric components. In parametric component, time dependent term is known up to proportional constant. In purely nonparametric component, time dependent term is an unknown function, and time dependent term in smoothly nonparametric component is an unknown but smoothly function. As an estimation method of this model, we use the weighted least square estimation by Huffer and McKeague (1991). We provide an illustrative example as well as a simulation study that compares the performance of our method with the ordinary least square method.

• Ocean and Polar Research
• /
• v.30 no.4
• /
• pp.373-378
• /
• 2008

Long-term tide-gauge data from around the Korean Peninsula were reanalyzed. Both the coastal water and the open sea surrounding the Korean Peninsula appeared to have been influenced by global warming. The long-term change in relative sea levels obtained from tidal stations showed a general rising trend, especially near Jeju Island. It is proposed that global warming may have caused shifting of the path of the Kuroshio branch (Tsushima Warm Current) toward Jeju Island, causing a persistent increase in the water levels along the coast of the island over the last few decades.

• Water Engineering Research
• /
• v.5 no.1
• /
• pp.1-16
• /
• 2004

The purpose of this study is to understand the characteristics of hourly PET(Potential Evapo Transpiration) variation estimated using Penman ET model. The estimated PET using Penman model was compared with measured ET. For this study, two subwatersheds were selected, and fluxes, meteorological data and soil moisture data were measured during the summer and winter days. During the winter days, the aerodynamic term of Penman ET is much greater than that of energy term of Penman ET for dry soil condition. The opposite phenomena appeared fer wet soil condition. During the summer days, energy term is much more important factor for ET estimation compared with aerodynamic term regardless of soil moisture condition. Penman ET, measured ET, and energy term show the similar hourly variation pattern mainly because the influence of net radiation on the estimation of Penman ET is much more significant compared with other variables. Even though there are much more soil moisture in the soil during the wet days, the estimated hourly ET from Penman model and measured hourly ET have smaller values compared with those of dry days, indicating the effect of cloudy weather condition.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2003.05a
• /
• pp.481-486
• /
• 2003

This paper presents experimental study on long-term behavior of RC and SRC columns. Shrinkage and creep tests were performed for two types of concrete used in manufacturing RC and SRC column specimens. Also, long-term tests under sustained load were carried out for RC and SRC column specimens. Shrinkage functions and creep coefficients to optimally fit the corresponding data were obtained from regression analysis and the regression results, ACI and CEB- FIP 90 model were applied to analyse long-term behvavior of RC and SRC column specimens. Creep coefficients calculated from test data were lower than those predicted by ACI and CEB-FIP 90 models. Long-term analysis results for RC and SRC column specimens using the regression results were relatively more accurate than those obtained using the existing models.

• Journal of the Korean Wood Science and Technology
• /
• v.17 no.4
• /
• pp.35-43
• /
• 1989

Nailed joints, which are commonly used in Wooden structures, transmit loads from one member to another and induce partial composite actions between members. Long-term loads induce creep slip in nailed joints and affect load sharing and partial composite action, which may reduce joint stiffness. Two theoretical viscous-viscoelastic models were developed for nailed joints to predict creep behavior under long-term variable loads. Those models were also used to predict stiffness changes under long-term variable loads. The stiffness of nailed joint is defined as a Secant modulus which is called the joint modulus or slip modulus. Input data for the models are the results of constant load tests under three different load levels. To verify the models, nailed joints were also tested under two long-term variable load functions. The predictions of the models were very close to the experimental data. Therefore, the theoretical viscous-viscoelastic models and procedures developed in this study can be applied to predict creep slip and the changes in joint moduli of nailed joints under long-term variable loads.

• Journal of Haehwa Medicine
• /
• v.27 no.1
• /
• pp.1-8
• /
• 2018

Objectives : The purpose of this study was to analyze the status of inpatients in long-term care hospital and the satisfaction of Korean medical treatment. Methods : These analyzed data were collected from patients who were hospitalized in long-term care hospital in Sejong from Jan. 1. 2017 to Dec. 31. 2017. To analyze the status and the satisfaction of inpatients, we calculated the data of inpatients and conducted a survey. Results : Preference for acupuncture treatment was 95.6% and preference for numbers of Korean medical treatment more than two times per week was 40%. In addition, confidence in Korean medical treatment was 4.31, improvement after Korean medical treatment was 4.07 and general satisfaction was 4.58 with five-point scale. Conclusions : Most of the patients in long-term care hospital were satisfied with Korean medical treatment.

• Journal of Korean Society for Atmospheric Environment
• /
• v.4 no.2
• /
• pp.67-71
• /
• 1988

To estimate the probability of short term concentration of air pollution using long term arithmetic average concentration, the procedure was developed and added to Texas Climatological Model version 2. In the procedure, such statistical characteristics that frequency distribution of short term concentration may be approximated by a lognormal distribution, were applied. This procedure is capable of estimating not only highest concentration for a variety of averaging times but also concentrations for arbitrary occurrence frequency. Evaluation of the procedure with the results of short term concentrations calculated by Texas Episodic Model version 8 using the meteorological data and emission data in Seoul shows that the procedure estimates concentrations fairly well for wide range of percentiles.