• Communications for Statistical Applications and Methods
• /
• v.21 no.2
• /
• pp.169-181
• /
• 2014

Marginalized random effects models (MREM) are commonly used to analyze longitudinal categorical data when the population-averaged effects is of interest. In these models, random effects are used to explain both subject and time variations. The estimation of the random effects covariance matrix is not simple in MREM because of the high dimension and the positive definiteness. A relatively simple structure for the correlation is assumed such as a homogeneous AR(1) structure; however, it is too strong of an assumption. In consequence, the estimates of the fixed effects can be biased. To avoid this problem, we introduce one approach to explain a heterogenous random effects covariance matrix using a modified Cholesky decomposition. The approach results in parameters that can be easily modeled without concern that the resulting estimator will not be positive definite. The interpretation of the parameters is sensible. We analyze metabolic syndrome data from a Korean Genomic Epidemiology Study using this method.

• Structural Engineering and Mechanics
• /
• v.55 no.3
• /
• pp.461-472
• /
• 2015

The existence of initial imperfections in manufacturing or assembly of double-layer space structures having hundreds or thousands of members is inevitable. Many of the imperfections, such as the initial curvature of the members and residual stresses in members, are all random in nature. In this paper, the probabilistic effect of initial curvature imperfections in the load bearing capacity of double-layer grid space structures with different types of supports have been investigated. First, for the initial curvature imperfection of each member, a random number is generated from a gamma distribution. Then, by employing the same probabilistic model, the imperfections are randomly distributed amongst the members of the structure. Afterwards, the collapse behavior and the ultimate bearing capacity of the structure are determined by using nonlinear push down analysis and this procedure is frequently repeated. Ultimately, based on the maximum values of bearing capacity acquired from the analysis of different samples, structure's reliability is obtained by using Monte Carlo simulation method. The results show the sensitivity of the collapse behavior of double-layer grid space structures to the random distribution of initial imperfections and supports type.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2004.05b
• /
• pp.682-684
• /
• 2004

The effect of 1/f noise by the random telegraph signal(RTS) on the phase noise and the jitter of CMOS ring Oscillator is investigated. 10 parallel piece-wise-linear current sources connected to each node model the RTS signals. The In, the power spectral density and the jitter of output of the ring oscillator are simulated as functions of the amplitude and time constant of RTS current source. It is confirmed that the increase of amplitude of RTS is directly related to the increase of the width of phase noise md the value of jitter. The shorter the time constant is, the wider width of FET peak and the larger value of cycle to cycle jitter are.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.5
• /
• pp.358-365
• /
• 2016

We propose a novel application of random forest, a machine learning-based general classification algorithm, to analyze the influence of design attributes on the silicon-to-SPICE (S2S) gap. To improve modeling accuracy, we introduce magnification of learning data as well as randomization for the counting of design attributes to be used for each tree in the forest. From the automatically generated decision trees, we can extract the so-called importance and impact indices, which identify the most significant design attributes determining the S2S gap. We apply the proposed method to actual silicon data, and observe that the identified design attributes show a clear trend in the S2S gap. We finally unveil 10nm key fin-shaped field effect transistor (FinFET) structures that result in a large S2S gap using the measurement data from 10nm test vehicles specialized for model-hardware correlation.

• Architectural research
• /
• v.8 no.2
• /
• pp.27-36
• /
• 2006

Structural uncertainties are generally modeled using probabilistic approaches in order to quantify uncertainties in behaviors of structures. This uncertainty results from the uncertainties of structural parameters. Monte Carlo methods have been usually carried out for analyses of uncertainty problems where no analytical expression is available for the forward relationship between data and model parameters. In such cases any direct mathematical treatment is impossible, however the forward relation materializes itself as an algorithm allowing data to be calculated for any given model. This study addresses a new method which is utilized as a basis for the uncertainty estimates of structural responses. It applies double uniform random numbers (i.e. DURN technique) to conventional Monte Carlo algorithm. In DURN method, the scenarios of uncertainties are sequentially selected and executed in its simulation. Numerical examples demonstrate the beneficial effect that the technique can increase uncertainty degree of structural properties with maintaining structural stability and safety up to the limit point of a breakdown of structural systems.

• Communications for Statistical Applications and Methods
• /
• v.16 no.6
• /
• pp.1031-1036
• /
• 2009

Under recent dramatic declines in response rates, various procedures have been considered among survey practitioners to reduce nonresponse in order to avoid its potential impairment to the inference. In the random digit dialing telephone surveys, substantial efforts are often required to obtain the initial contact for the screener interview. To reduce a burden with higher data collection costs, refusal conversion can be administered only to a random portion of the sample, reducing nonresponse (bias) with an expense of sample variability increment due to the associated weight adjustment. In this paper, we provide ways to determine the optimal subsampling rate using a linear cost model. Our approach for refusal subsampling is to predetermine a random portion from the full sample and to apply refusal conversion efforts if needed only to the subsample.

• Communications for Statistical Applications and Methods
• /
• v.27 no.5
• /
• pp.523-533
• /
• 2020

This paper suggests three available methods for finding nonnegative estimates of variance components of the random effects in mixed models. The three proposed methods based on the concepts of projections are called projection method I, II, and III. Each method derives sums of squares uniquely based on its own method of projections. All the sums of squares in quadratic forms are calculated as the squared lengths of projections of an observation vector; therefore, there is discussion on the decomposition of the observation vector into the sum of orthogonal projections for establishing a projection model. The projection model in matrix form is constructed by ascertaining the orthogonal projections defined on vector subspaces. Nonnegative estimates are then obtained by the projection model where all the coefficient matrices of the effects in the model are orthogonal to each other. Each method provides its own system of linear equations in a different way for the estimation of variance components; however, the estimates are given as the same regardless of the methods, whichever is used. Hartley's synthesis is used as a method for finding the coefficients of variance components.

• The Journal of Industrial Distribution & Business
• /
• v.10 no.11
• /
• pp.15-23
• /
• 2019

Purpose: Global economic integration has provided good opportunities and conditions for the development of foreign direct investment in Finances. Therefore, this paper attempts to explore what determines foreign direct investment in Finances of Organization for Economic Co-operation and Development (OECD) countries. Research design, data and methodology: This paper employs the panel data over the period 2005-2017 and uses the random effect model to estimate this proposition. Results: The results indicate that the foreign direct investment in services, growth rate of GDP, interest rate and saving are positively related with foreign direct investment in finances. Conversely, the growth rate of wage and fluctuation rate of exchange rate are negatively related with foreign direct investment in finances. Moreover, the results verify that the effect of these variables on foreign direct investment in finances is different before and after 2008 (global economic crisis). In addition, the results also manifest that the regional effect exists. Namely, the effect of these variables on foreign direct investment in finances between G7 countries and G20 countries exist significant difference. Conclusions: Those variables used in this paper are related with foreign direct investment in Finances of (OECD) countries.

• Journal of Korean Society of Transportation
• /
• v.32 no.6
• /
• pp.662-674
• /
• 2014

In this study, accident frequency prediction models were constructed by collecting variables such as geometric structures, safety facilities, traffic volume and weather conditions, land use, highway design-satisfaction criteria along 780km (4,372 sections) of 4 lane-highways over 8 areas. As for models, a fixed parameter model and a random parameter model were employed. In the random parameter model, some influences were reversed as the range was expressed based on specific probability in the case of no fixed coefficients. In the fixed parameter model, the influences of independent variables on accident frequency were interpreted by using one coefficient, but in the random parameter model, more various interpretations were took place. In particular, curve radius, securement of shoulder lane, vertical grade design criteria satisfaction showed both positive and negative influence, according to specific probability. This means that there could be a reverse effect depending on the behavioral characteristics of drivers and the characteristics of highway sections. Rather, they influence the increase of accident frequency through the all sections.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.05a
• /
• pp.79-84
• /
• 2003

Likelihood estimation in random-effect models is often complicated because the marginal likelihood involves an analytically intractable integral. Numerical integration such as Gauss-Hermite quadrature is an option, but is generally not recommended when the dimensionality of the integral is high. An alternative is the use of hierarchical likelihood, which avoids such burdensome numerical integration. These two approaches for fitting binary data are compared and the advantages of using the hierarchical likelihood are discussed. Random-effect models for binary outcomes and for bivariate binary-continuous outcomes are considered.