• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.335-342
• /
• 2015

Joint hierarchical generalized linear models proposed by Molas et al. (2013) extend the simple longitudinal model into multiple models fitted jointly. It can easily handle the correlation of multivariate longitudinal data. In this paper, we apply this method to analyze KoGES cohort dataset. Fixed unknown parameters, random effects and variance components are estimated based on a standard framework of h-likelihood theory. Furthermore, based on the conditional Akaike information criterion the correlated covariance structure of random-effect model is selected rather than an independent structure.

• Earthquakes and Structures
• /
• v.22 no.3
• /
• pp.219-230
• /
• 2022

Knowing that the variability of soil properties is an important source of uncertainty in geotechnical analyses, we will study in this paper the effect of this variability on the seismic response of a structure within the framework of Soil Structure Interaction (SSI). We use the proposed and developed model (N2-ISS, Mekki et al., 2014). This approach is based on an extension of the N2 method by determining the capacity curve of the fixed base system oscillating mainly in the first mode, then modified to obtain the capacity curve of the system on a flexible basis using the concept of the equivalent nonlinear oscillator. The properties of the soil that we are interested in this paper will be the shear wave velocity and the soil damping. These parameters will be modeled at first, as independent random fields, then, the two parameters will be correlated. The results obtained showed the importance of the use of random field in the study of SSI systems. The variability of soil damping and shear wave velocity introduces significant uncertainty not only in the evaluation of the damping of the soil-structure system but also in the estimation of the displacement of the structure and the base-shear force.

• Computational Structural Engineering : An International Journal
• /
• v.1 no.2
• /
• pp.139-150
• /
• 2001

An initial approach for the identification of physical causes underlying the spatial coherency of seismic ground motions it presented. The approach relies on the observation that amplitude and phase variability of seismic data recorded over extended areas around the amplitude and phase of a common, coherent component are correlated. It suffices then to examine the physical causes for the amplitude variability in the seismic motions, in order to recognize the causes for the phase variability and, consequently, the spatial coherency. In this study, the effect of randomness in the shear wave velocity at a site on the amplitude variability of the surface motions mi investigated by means of simulations. The amplitude variability of the simulated motions around the amplitude of the common component is contained within envelope functions, the shape of which suggests, on a preliminary basis, the trend of the decay of coherency with frequency.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.2
• /
• pp.385-392
• /
• 2012

Logistic regression is a well known binary classification method in the field of statistical learning. Mixed-effect regression models are widely used for the analysis of correlated data such as those found in longitudinal studies. We consider kernel extensions with semiparametric fixed effects and parametric random effects for the logistic regression. The estimation is performed through the penalized likelihood method based on kernel trick, and our focus is on the efficient computation and the effective hyperparameter selection. For the selection of optimal hyperparameters, cross-validation techniques are employed. Numerical results are then presented to indicate the performance of the proposed procedure.

• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.353-360
• /
• 2015

Various statistical models have been proposed over the last decade for spatially correlated Gaussian outcomes. The spatial linear mixed model (SLMM), which incorporates a spatial effect as a random component to the linear model, is the one of the most widely used approaches in various application contexts. Employing link functions, SLMM can be naturally extended to spatial generalized linear mixed model for non-Gaussian outcomes (SGLMM). We review popular SGLMMs on non-Gaussian spatial outcomes and demonstrate their applications with available public data.

• Asian-Australasian Journal of Animal Sciences
• /
• v.19 no.11
• /
• pp.1524-1528
• /
• 2006

A total of 437 animals, comprising lambs aged between 3 and 12 months and adults of either sex of Muzaffarnagari sheep maintained at the Central Institute for Research on Goats, Makhdoom, Farah, Mathura, India were screened to assess the prevalence of Haemonchus contortus infection following natural infection and to identify the various factors affecting faecal egg count (FEC) and packed cell volume (PCV) of ewes and their genetic control. The relationships between FEC, PCV and body weight were also estimated. The prevalence rate for H. contortus infection in the flock under study was 15.7% indicating much lower occurrence of worm infection in lambs up to one year of age. On the other hand, a large proportion i.e., 67.7% of sheep was refractive to natural H. contortus infection. The random effect of sire significantly contributed (p<0.01) variation in log-transformed FEC (LFEC) of ewes. The season of birth had a significant (p<0.01) effect on LFEC of ewes. The lactating ewes had significantly (p<0.01) higher faecal egg counts compared to dry and pregnant ewes. The linear regression effects of the age of ewes on LFEC of animals were significant (p<0.01) in the present study. The heritabilities of LFEC, PCV and body weights of ewes during the course of infection were moderate to high in magnitude and ranged from 0.24 to 0.47. The LFEC of ewes was significantly (p<0.05) and negatively correlated with PCV at both genetic and phenotypic level. The genetic and phenotypic relationships between LFEC and body weights of ewes were -0.26 and -0.06 for this breed. The genetic correlation of PCV and body weight of ewes was positive and high (0.58) and statistically significant (p<0.05) but it was negatively correlated (-0.01) with body weight at the phenotypic level.

• Journal of the Korean Society of Food Culture
• /
• v.7 no.1
• /
• pp.1-8
• /
• 1992

This study was carried out to evaluate Korean eating behavior which is highly correlated with their nutritional status, and to analyze the effect of various factors on eating behavior. The above information was used to develop a nutritional status for Korea. The 2000 Korean people were selected with the stratified random sampling method. This study used a questionaire as instrument tool. The questionaire consists of :1) socio-demographic characteristics of the subjects; 2) the valuation of food and nutrition; 3) the concern of food and nutrition; 4) psychological health condition; 5) physical health condition; 6) nutrient consumption status and 7) analysis of eating behavior. Data were analyzed by using a SPSS PC Package. Significant differences and correlation among variables were determined by the t-test, $x^2-test$, ANOVA, pearson's correlation coefficient and Multiple regression analysis. The results of this study can be summarized as follow, All nutrient intakes were significantly correlated with eating behavior score(p<0.001). Factors such as socio-economic status, valuation and concern on food and nutrition, and psychological health condition had significant relationship with eating behavior. But the physical health condition had no significant effect on it. Multiple regression analysis showed that valuation of food and nutrition made the greatest contribution(35.6% explained) and concern made the second greatest contribution(10.5% explained). The third was education level(9.8% explained), and the forth psychological health condition(1.8% explained).

• Asian-Australasian Journal of Animal Sciences
• /
• v.15 no.12
• /
• pp.1690-1694
• /
• 2002

Genetic parameters were estimated for birth weight and weaning weight from three year (1991-1993) data totalling 1100 records of 25 rams to 205 ewes of Indigenous Sabi flock maintained at Grasslands Research Station in Zimbabwe. AIREML procedures were used fitting an Animal Model. The statistical model included the fixed effects of year of lambing, sex of lamb, birth type and the random effect of ewe. Weight of ewe when first joined with ram was included as a covariate. Direct heritability estimates of 0.27 and 0.38, and maternal heritability estimates of 0.24 and 0.09, were obtained for birth weight and weaning weight, respectively. The total heritability estimates were 0.69 and 0.77 for birth weight and weaning weight, respectively. Direct-aternal genetic correlations were high and positive. The corresponding genetic covariance estimates between direct and maternal effects were positive and low, 0.25 and 0.18 for birth weight and weaning weight, respectively. Responses to selection were 0.8 kg and 0.14 kg for birth weight and weaning weight, respectively. The estimated expected correlated response to selection for birth weight by directly selecting for weaning weight was 0.26. Direct heritabilities were moderate; as a result selection for any of these traits should be successful. Maternal heritabilities were low for weaning weight and should have less effect on selection response. Indirect selection can give lower response than direct selection.

• Communications for Statistical Applications and Methods
• /
• v.27 no.4
• /
• pp.487-500
• /
• 2020

Insurers face increasing demands for cyber liability; entailed in part by a variety of new forms of risk of data breaches. As data breach occurrences develop, our understanding of the volatility in data breach counts has also become important as well as its expected occurrences. Volatility clustering, the tendency of large changes in a random variable to cluster together in time, are frequently observed in many financial asset prices, asset returns, and it is questioned whether the volatility of data breach occurrences are also clustered in time. We now present volatility analysis based on INGARCH models, i.e., integer-valued generalized autoregressive conditional heteroskedasticity time series model for frequency counts due to data breaches. Using the INGARCH(1, 1) model with data breach samples, we show evidence of temporal volatility clustering for data breaches. In addition, we present that the firms' volatilities are correlated between some they belong to and that such a clustering effect remains even after excluding the effect of financial covariates such as the VIX and the stock return of S&P500 that have their own volatility clustering.

• Journal of Preventive Medicine and Public Health
• /
• v.37 no.4
• /
• pp.381-389
• /
• 2004

Objective : To compare the efficacy between SKI306X and Diclofenac by using generalized estimating equations (GEE) methodology in the analysis of correlated bivariate binary outcome data in Osteoarthritis (OA) diseases. Methods : A randomized, double-blind, active comparator-controlled, non-inferiority clinical trial was conducted at 5 institutions in Korea with the random assignment of 248 patients aged 35 to 75 years old with OA of the knee and clinical evidence of OA. Patients were enrolled in this study if they had at least moderate pain in the affected knee joint and a score larger than 35mm as assessed by VAS (Visual Analog Scale). The main exposure variable was treatment (SKI 306X vs. Diclofenac) and other covariates were age, sex, BMI, baseline VAS, center, operation history (Yes/No), NSAIDS (Y/N), acupuncture (Y/N), herbal medicine (Y/N), past history of musculoskeletal disease (Y/N), and previous therapy related with OA (Y/N). The main study outcome was the change of VAS pain scores from baseline to the 2nd and 4th weeks after treatment. Pain scores were obtained as baseline, 2nd and 4th weeks after treatment. We applied GEE approach with empirical covariance matrix and independent(or exchangeable) working correlation matrix to evaluate the relation of several risk factors to the change of VAS pain scores with correlated binary bivariate outcomes. Results : While baseline VAS, age, and acupuncture variables had protective effects for reducing the OA pain, its treatment (Joins/Diclofenac) was not statistically significant through GEE methodology (ITT:aOR=1.37, 95% CI=(0.8200, 2.26), PP:aOR=1.47, 95% CI=(0.73, 2.95)). The goodness-of-fit statistic for GEE (6.55, p=0.68) was computed to assess the adequacy of the fitted final model. Conclusions : Both ANCOVA and GEE methods yielded non statistical significance in the evaluation of non-inferiority of the efficacy between SKI306X and Diclofenac. While VAS outcome for each visit was applied in GEE, only VAS outcome for the fourth visit was applied in ANCOVA. So the GEE methodology is more accurate for the analysis of correlated outcomes.