• Communications for Statistical Applications and Methods
• /
• 제24권4호
• /
• pp.383-396
• /
• 2017

The model in our approach assumes that computer responses are a realization of a Gaussian processes superimposed on a regression model called a Gaussian process regression model (GPRM). Selecting a subset of variables or building a good reduced model in classical regression is an important process to identify variables influential to responses and for further analysis such as prediction or classification. One reason to select some variables in the prediction aspect is to prevent the over-fitting or under-fitting to data. The same reasoning and approach can be applicable to GPRM. However, only a few works on the variable selection in GPRM were done. In this paper, we propose a new algorithm to build a good prediction model among some GPRMs. It is a post-work of the algorithm that includes the Welch method suggested by previous researchers. The proposed algorithms select some non-zero regression coefficients (${\beta}^{\prime}s$) using forward and backward methods along with the Lasso guided approach. During this process, the fixed were covariance parameters (${\theta}^{\prime}s$) that were pre-selected by the Welch algorithm. We illustrated the superiority of our proposed models over the Welch method and non-selection models using four test functions and one real data example. Future extensions are also discussed.

• Science of Emotion and Sensibility
• /
• 제7권3호
• /
• pp.15-25
• /
• 2004

The regression analysis and Adaptive -Network based Fuzzy-inference system (ANFIS) were applied to the explanation on human's visual texture of cotton fabrics with 7 mechanical properties. The ANFIS uses the structure with fuzzy membership function and neural network. The results obtained by the statistical analysis through the coefficient of correlation and regression analysis showed that subjective texture had a linear relationship with mechanical properties. But It had a relatively low coefficient of determination and was difficult that the statistical analysis explained other relationship with the exception of a lineality and interaction among mechanical properties. Comparing the statistical analysis, the ANFIS was an effective tool to explain human's non-linear perceptions and their interactions. But to apply ANFIS to human's perceptions more effectively, it is necessary to discriminate effective input variables through controlling the properties of samples.

• Journal of Korean Society of Transportation
• /
• 제31권1호
• /
• pp.65-76
• /
• 2013

Traffic accidents at signalized intersections have been increased annually so that it is required to examine the causation to reduce the accidents. However, the current existing accident models were developed mainly by using non-linear regression models such as Poisson methods. These non-linear regression methods lack to reveal the complicated causation for traffic accidents, though they are the right choice to study randomness and non-linearity of accidents. Therefore, it is required to utilize another statistical method to make up for the lack of the non-linear regression methods. This study developed accident prediction models for 4 legged signalized intersections with Poisson methods and compared them with structural equation models. This study used structural equation methods to reveal the complicated causation of traffic accidents, because the structural equation method has merits to explain more causational factors for accidents than others.

• Journal of Electrical Engineering and Technology
• /
• 제11권6호
• /
• pp.1628-1633
• /
• 2016

This paper derives an effective shape of the ferromagnetic pole pieces (low-speed rotor) for the reduction of transmission torque ripple in a magnetic-geared machine based on a Box-Behnken design (BBD). In particular, using a non-linear finite element method (FEM) based on 2-D numerical analysis, we conduct a numerical investigation and analysis between independent variables (selected by the BBD) and reaction variables. In addition, we derive a regression equation for reaction variables according to the independent variables by using multiple regression analysis and analysis of variance (ANOVA). We assess the validity of the optimized design by comparing characteristics of the optimized model derived from a response surface analysis and an initial model.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 한국근적외분광분석학회 2001년도 NIR-2001
• /
• pp.1524-1524
• /
• 2001

A previous paper(Ito et al., 2000) has described the improvement of the standard error(SEC and SEP) of the predicted soluble solids(Brix) in a melon cultivar by non-contact mode with a fiber optic probe. Then we examined the immature and mature fruits. The objective of this study was to determine if non-contact mode could improve the standard error of the predicted Brix of matured melon fruits from cross progeny as well as the contact mode(usual method). The optical absorption spectrum was measured using a NIR Systems model 6500 spectrophotometer. A commercial spectral program(NSAS ver. 3.27) was used for multiple linear regression analysis. Absorbances of 902 and in the vicinity of 877 nm were included as the independent variables in both multiple regression equations. These wavelengths are key wavelengths for non-destructive Brix determination. When the results for the contact mode and non-contact mode are compared, the latter mode improved the former standard error(SEP and RMS).

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제16권12호
• /
• pp.8791-8799
• /
• 2015

In this paper, One-way car-sharing System represents the verification system consider events in revenue effects. Revenue which the time and distance represented the graph, compare one-way vehicle relocate car-sharing system which proven in existing international papers with one-way vehicle relocate car-sharing system consider the event currently in the Korea. Especially, The maximum profit according to the distance and time were assessed through multiple linear regression analysis, and there are probable maximum loss allow for the maximum loss. The company suggested using the event as a discount coupon to customers through various marketing strategies, and then focused on increasing customer demand. So, Correlation analysis to determine the maximum revenue of the actual travel distance and time were carried out through Non-linear Regression.

• Asia-Pacific Journal of Business
• /
• 제14권4호
• /
• pp.233-254
• /
• 2023

Purpose - This study aims to verify whether the effect of tax avoidance on corporate value is non-linear in the Korean financial markets. Design/methodology/approach - This study believes that the cause of the inconsistent empirical analysis results of previous studies that verified the relationship between tax avoidance and firm value may be an error in assuming linearity, and verifies whether a nonlinear relationship exists. The sample company in this study is a December settlement corporation listed on the Korean stock market, and the analysis period is from 2000 to 2021. In the empirical analysis model, Tobin's Q is used as a proxy for corporate value, tax avoidance is used as the main independent variable, and a regression model is designed with corporate size, growth rate, and debt ratio set as control variables. Findings - As a result of the empirical analysis, it can be confirmed that there is an inverted U-shaped nonlinear relationship between tax avoidance and corporate value. In the additional analysis using Ohlson (1995) firm valuation model for the robustness of the results of the empirical analysis, the same nonlinear value relationship between tax avoidance can be confirmed. Research implications or Originality - This study is considered to be meaningful in that it verifies the non-linear relationship of tax avoidance, which has not been attempted in previous studies. The meaning of the inverted U-shaped nonlinear relationship presented in this study is that corporate tax avoidance acts as a factor that increases corporate value up to a certain level, but rather becomes a factor that decreases corporate value when it exceeds a critical point. These results are expected to provide new perspectives and perspectives on tax avoidance to companies belonging to the Korean capital market.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제5권2호
• /
• pp.277-286
• /
• 2013

Cross-shore sediment transport is very important factor in the design of coastal structures, and the beach profile is mainly affected by a number of parameters, such as wave height and period, beach slope, and the material properties of the bed. In this study cross-shore sediment movement was investigated using a physical model and various offshore bar geometric parameters were determined by the resultant erosion profile. The experiments on cross- shore sediment transport carried out in a laboratory wave channel for initial base slopes of 1/8, 1/10 and 1/15. Using the regular waves with different deep-water wave steepness generated by a pedal-type wave generator, the geometrical of sediment transport rate and considerable characteristics of beach profiles under storm conditions and bar parameters affecting on-off shore sediment transport are investigated for the beach materials with the medium diameter of $d_{50}$=0.25, 0.32, 0.45, 0.62 and 0.80 mm. Non-dimensional equations were obtained by using linear and non-linear regression methods through the experimental data and were compared with previously developed equations in the literature. The results have shown that the experimental data fitted well to the proposed equations with respect to the previously developed equations.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• 제47권1호
• /
• pp.41-50
• /
• 2024

Korea is facing a significant problem with historically low fertility rates, which is becoming a major social issue affecting the economy, labor force, and national security. This study analyzes the factors contributing to the regional gap in fertility rates and derives policy implications. The government and local authorities are implementing a range of policies to address the issue of low fertility. To establish an effective strategy, it is essential to identify the primary factors that contribute to regional disparities. This study identifies these factors and explores policy implications through machine learning and explainable artificial intelligence. The study also examines the influence of media and public opinion on childbirth in Korea by incorporating news and online community sentiment, as well as sentiment fear indices, as independent variables. To establish the relationship between regional fertility rates and factors, the study employs four machine learning models: multiple linear regression, XGBoost, Random Forest, and Support Vector Regression. Support Vector Regression, XGBoost, and Random Forest significantly outperform linear regression, highlighting the importance of machine learning models in explaining non-linear relationships with numerous variables. A factor analysis using SHAP is then conducted. The unemployment rate, Regional Gross Domestic Product per Capita, Women's Participation in Economic Activities, Number of Crimes Committed, Average Age of First Marriage, and Private Education Expenses significantly impact regional fertility rates. However, the degree of impact of the factors affecting fertility may vary by region, suggesting the need for policies tailored to the characteristics of each region, not just an overall ranking of factors.

• Korean Journal of Exercise Nutrition
• /
• 제25권1호
• /
• pp.23-29
• /
• 2021

[Purpose] This preliminary study aimed to develop a regression model to estimate the non-exercise activity thermogenesis (NEAT) of Korean adults using various easy-to-measure dependent variables. [Methods] NEAT was measured in 71 healthy adults (male n = 29; female n = 42). Statistical analysis was performed to develop a NEAT estimation regression model using the stepwise regression method. [Results] We confirmed that age^A, weight^B, heart rate (HR)_average^C, weight × HR_average^D, weight × HR_sum^E, systolic blood pressure (SBP) × HR_rest^F, fat mass ÷ height^2G, gender × HR_average^H, and gender × weight × HR_sum^I were important variables in various NEAT activity regression models. There was no significant difference between the measured NEAT values obtained using a metabolic gas analyzer and the predicted NEAT. [Conclusion] This preliminary study developed a regression model to estimate the NEAT in healthy Korean adults. The regression model was as follows: sitting = 1.431 - 0.013 × (A) + 0.00014 × (D) - 0.00005 × (F) + 0.006 × (H); leg jiggling = 1.102 - 0.011 × (A) + 0.013 × (B) + 0.005 × (H); standing = 1.713 - 0.013 × (A) + 0.0000017 × (I); 4.5 km/h walking = 0.864 + 0.035 × (B) + 0.0000041 × (E); 6.0 km/h walking = 4.029 - 0.024 × (C) + 0.00071 × (D); climbing up 1 stair = 1.308 - 0.016 × (A) + 0.00035 × (D) - 0.000085 × (F) - 0.098 × (G); and climbing up 2 stairs = 1.442 - 0.023 × (A) - 0.000093 × (F) - 0.121 × (G) + 0.0000624 × (E).