• The Korean Journal of Applied Statistics
• /
• v.33 no.1
• /
• pp.25-46
• /
• 2020

This paper presents ordinal probit semiparametric regression models using Bayesian Spectral Analysis Regression (BSAR) method. Ordinal probit regression is a way of modeling ordinal responses - usually more than two categories - by connecting the probability of falling into each category explained by a combination of available covariates using a probit (an inverse function of normal cumulative distribution function) link. The Bayesian probit model facilitates posterior sampling by bringing a latent variable following normal distribution, therefore, the responses are categorized by the cut-off points according to values of latent variables. In this paper, we extend the latent variable approach to a semiparametric model for the Bayesian ordinal probit regression with nonparametric functions using a spectral representation of Gaussian processes based BSAR method. The latent variable is decomposed into a parametric component and a nonparametric component with or without a shape constraint for modeling ordinal responses and predicting outcomes more flexibly. We illustrate the proposed methods with simulation studies in comparison with existing methods and real data analysis applied to a Korean National Health and Nutrition Examination Survey (KNHANES) 2016 for investigating nonparametric relationship between smoking behavior and coffee intake.

• Journal of the Korean Society for Library and Information Science
• /
• v.57 no.1
• /
• pp.45-68
• /
• 2023

The purpose of this study is to longitudinally analyze impact factors on library use. One of library use indicators, the number of circulated books was statistically analyzed with latent growth model and growth mixture model. Library data from 2014 to 2019 were collected from the National Library Statistics System, and 846 public libraries were analyzed. As results, the number of circulated books were decreased, but it was tempered. Next, with controlling the factor affecting the dependent variable, the size of collection and the number of participants in reading programs provided by public libraries were statistically significant. Lastly, 5 classes were identified by applying the growth mixture model, and the number of librarians was significantly associated with trajectory class membership.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.21 no.6
• /
• pp.175-182
• /
• 2013

Lately, customers think highly of the emotional satisfaction and as a result, issues on odor are matters of concern. The cases are odor of interior material and air-conditioner of vehicles. In particualar, with respect to the odor of air-conditioner, customers strongly claimed defects with provocative comments : "It smells like something rotten," "It smells like a foot odor," "It stinks like a rag." Generally, it is known that mold of evaporator core in the air-conditioning system decays and this produce VOCs which causes the odor to occur. In this study, partial least squares regression model is applied to predict the strength of the odor and select of important VOCs which affect car air conditioning smell. The PLS method is basically a particular multilinear regression algorithm which can handle correlated inputs and limited data. The number of latent variable is determined by the point which is stabilized mean absolute deviations of VOCs data. Also multiple linear regression is carried out to confirm the validity of PLS method.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.367-374
• /
• 2021

Stock price prediction is a subject of research in various fields such as economy, statistics, computer engineering, etc. In recent years, researches on predicting the movement of stock prices by learning artificial intelligence models from various indicators such as basic indicators and technical indicators have become active. This study proposes a deep learning model that predicts the ups and downs of KOSPI from overseas indices such as S&P500, past KOSPI indices, and trading trends by KOSPI investors. The proposed model extracts a latent variable using a stacked auto-encoder to predict stock price fluctuations, and predicts the fluctuation of the closing price compared to the market price of the day by learning an LSTM suitable for learning time series data from the extracted latent variable to decide to buy or sell based on the value. As a result of comparing the returns and prediction accuracy of the proposed model and the comparative models, the proposed model showed better performance than the comparative models.

• Journal of the Korean Regional Science Association
• /
• v.31 no.3
• /
• pp.19-37
• /
• 2015

The purpose of this study is to confirm a structural relationship on factors affecting ratio of public transportation spending to a car-owning household's total transportation expenditure. For this purpose, informations of household's attributes and activities were gathered using the 13th Korean Labor and Income Panel Study (KLIPS), and information of land-use and transportation conditions on their residential locations was collected and processed. A structural equation model (SEM) on determinants affecting ratio of public transportation expenditure was constructed, based on an execution result of factor analysis using the analyzing database. The latent variables were derived as land-use/transportation characteristic, household's attribute and household's activity. In the analyzing result of the SEM, the entire latent variables were significant. And, the first two latent variables had positive influences, and the last latent variable had a negative impact. To promote public transportation use of the car-owning households, this study suggests that the policies such as enhancement of convenience in public transportation use for the household's activities and improvement of the land-use/transport conditions are required.

• Journal of the Korean Society of Safety
• /
• v.36 no.3
• /
• pp.60-65
• /
• 2021

Human error is often in part in the cause of accidents and the result of various factors in an organization. Accidents should be investigated to elucidate all causes. Therefore, to reduce accidents, it is necessary to identify which factors affect human error within the organization. In this study, five groups of influencing factors on human error were selected using previousresearch, and operational definitions were made based on them. In addition, a questionnaire for measuring latent variables by operational definition was developed as an observation variable, and responses were received from employees of chemical companies in Ulsan. Based on SEM (structural equation modeling) analysis, 1) confirmatory factor analysis of variables in the human error model, 2) reliability and validity of latent variables, 3) correlations among latent variables, 4) influencing coefficients among influence factors, and 5) the verification results of the paths that these influencing factors have on human error are introduced in this study.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.36 no.4
• /
• pp.64-70
• /
• 2013

Estimation approaches for casual relation model with high-order factors have strict restrictions or limits. In the case of ML (Maximum Likelihood), a strong assumption which data must show a normal distribution is required and factors of exponentiation is impossible due to the uncertainty of factors. To overcome this limitation many PLS (Partial Least Squares) approaches are introduced to estimate the structural equation model including high-order factors. However, it is possible to yield biased estimates if there are some differences in the number of measurement variables connected to each latent variable. In addition, any approach does not exist to deal with general cases not having any measurement variable of high-order factors. This study compare several approaches including the repeated measures approach which are used to estimate the casual relation model including high-order factors by using PLS (Partial Least Squares), and suggest the best estimation approach. In other words, the study proposes the best approach through the research on the existing studies related to the casual relation model including high-order factors by using PLS and approach comparison using a virtual model.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.6 no.1
• /
• pp.83-87
• /
• 2012

This study is aimed to identify latent classes which are based the change patterns in assistive technology device use among worker with disabilities and to test the effects of independent variables(gender, education, disability type, disability density, activity and participation of ICF: ICF, subjective socioeconomic status: SES, job satisfaction, life satisfaction) on determining latents classes. This study applied Nagin's(1999) semi-parametric group based approach to the panel survey of employment for the disabled. Because dependant variable has dichotomous scale, logit model was used. The results identified three latent classes, which could be defined based on the patterns as follows; assistive device continued use group, assistive device mid-level use group, assistive device sharp decline use group. The effects of the independent variables on the latent classes was tested by multinomial logit analysis. The results showed that education, disability type, ICF, SES, and life satisfaction were significant determinants of the latent classes. Finally, the implications based on analysis results were suggested.

• International Journal of Reliability and Applications
• /
• v.6 no.2
• /
• pp.101-115
• /
• 2005

This article provides a new class of multivariate linear failure rate distributions where every component is a mixture of linear failure rate distribution. The new class includes several multivariate and bivariate models including Marslall and Olkin type. The approach in this paper is based on the introducing a linear failure rate distributed latent random variable. The distribution of minimum in a competing risk model is discussed.

• International Journal of Reliability and Applications
• /
• v.7 no.1
• /
• pp.55-69
• /
• 2006

This paper presents a new class of multivariate distributions with Pareto where dependence among the components is characterized by a latent random variable. The new class includes several multivariate and bivariate models of Marshall and Olkin type. It is found the bivariate distribution with Pareto is positively quadrant dependent and its mixture. Some important structural properties of the bivariate distributions with Pareto are discussed. The distribution of minimum in a competing risk Pareto model is derived.