• Communications for Statistical Applications and Methods
• /
• v.28 no.2
• /
• pp.189-204
• /
• 2021

In a regression analysis, a single best model is usually selected among several candidate models. However, it is often useful to combine several candidate models to achieve better performance, especially, in the prediction viewpoint. Model combining methods such as stacking and Bayesian model averaging (BMA) have been suggested from the perspective of averaging candidate models. When the candidate models include a true model, it is expected that BMA generally gives better performance than stacking. On the other hand, when candidate models do not include the true model, it is known that stacking outperforms BMA. Since stacking and BMA approaches have different properties, it is difficult to determine which method is more appropriate under other situations. In particular, it is not easy to find research papers that compare stacking and BMA when regression model assumptions are violated. Therefore, in the paper, we compare the performance among model averaging methods as well as a single best model in the linear regression analysis when standard linear regression assumptions are violated. Simulations were conducted to compare model averaging methods with the linear regression when data include outliers and data do not include them. We also compared them when data include errors from a non-normal distribution. The model averaging methods were applied to the water pollution data, which have a strong multicollinearity among variables. Simulation studies showed that the stacking method tends to give better performance than BMA or standard linear regression analysis (including the stepwise selection method) in the sense of risks (see (3.1)) or prediction error (see (3.2)) when typical linear regression assumptions are violated.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.56 no.4
• /
• pp.29-39
• /
• 2014

In this study, we performed a non-stationary frequency analysis using a power model and the model was applied for Seoul, Daegu, Daejeon, Mokpo sites in Korea to estimate the probable precipitation amount at the target years (2020, 2050, 2080). We used the annual maximum precipitation of 24 hours duration of precipitation using data from 1973 to 2009. We compared results to that of non-stationary analyses using the linear and logistic regression. The probable precipitation amounts using linear regression showed very large increase in the long term projection, while the logistic regression resulted in similar amounts for different target years because the logistic function converges before 2020. But the probable precipitation amount for the target years using a power model showed reasonable results suggesting that power model be able to reflect the increase of hydrologic extremes reasonably well.

• Fibers and Polymers
• /
• v.7 no.2
• /
• pp.203-212
• /
• 2006

The purpose of this study was to examine the effects of the structural properties of plain knit fabrics on the subjective perception of textures, sensibilities, and preference among consumers. This study, then, aimed to provide useful information with respect to planning and designing knitted fabrics by predicting the subjective characteristics analyzed according to their structural properties. For this purpose, we employed statistical analysis tools, such as factor and regression analysis and an adaptive-network-based fuzzy inference system(ANFIS), thereby combining the merits of fuzzy and neural networks and presupposing a non-linear relationship. Through factor analysis, we also categorized the subjective textures into 'roughness', 'softness', 'bulkiness' and 'stretch-ability' with R2=70.32%: and categorized the sensibilities into 'Stable/Neat', 'Natural/Comfortable' and 'Feminine/Elegant' with R2=68.12%. We analyzed subjective textures, sensibilities, and preference with ANFIS, assuming non-linear relationships; consequently, we were able to generate three or four fuzzy rules using wool/rayon fiber content and loop length as input data. The textures of roughness and softness exhibited a linear relationship, but other subjective characteristics demonstrated a non-linear input-output relationship. Compared with linear regression analysis, the ANFIS exhibited had higher predictive power with respect to predicting subjective characteristics.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.5
• /
• pp.851-856
• /
• 2016

Wind speed is heavily fluctuated and quite local than other weather elements. It is difficult to improve the accuracy of prediction only in a numerical prediction model. An MOS (Model Output Statistics) technique is used to correct the systematic errors of the model using a statistical data analysis. The Most of previous MOS has used a linear regression model for weather prediction, but it is hard to manage an irregular nature of prediction of wind speed. In order to solve the problem, a nonlinear regression method using SVR (Support Vector Regression) is introduced for a development of MOS for wind speed prediction. Experiments are performed for KLAPS (Korea Local Analysis and Prediction System) re-analysis data from 2007 to 2013 year for Jeju Island and Busan area in South Korea. The MLR and SVR based linear and nonlinear methods are compared to each other for prediction accuracy of wind speed. Also, the comparison experiments are executed for the variation in the number of UM elements.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.6
• /
• pp.371-378
• /
• 2003

The compression index of clay distributed in the west and south coast of the Korean Peninsula had been studied. Compression index was obtained from the conventional consolidation test, and was conducted accordingly to obtain the field virgin compression curve by means of Schmertmann's graphical correction. To examine a correlation closely between physical properties of soils($e_o$, LL, w) and compression index(Cc), linen. and non-linear regression analysis were employed based on the data collected from tests. The conclusions are as follows. The compression index obtained by means of Schmereann's graphical correction is about 1.16 times for the value of original oedometer test curve for U/D samples. Non-liner regression curve was preferable to establish a correlation equation rather than linear regression curve. All derived equations so far achieved have been summarized and given. However, linear equation is better for practical use so that part by part simplified linear equations were also suggested alternatively together with their own non-linear regression curve.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.5
• /
• pp.477-482
• /
• 2015

Linear regressions and evolutionary nonlinear regression based compensation techniques for the short-range prediction of wind speed are investigated. Development of an efficient MOS(Model Output Statistics) is necessary to correct systematic errors of the model, but a linear regression based MOS is hard to manage an irregular nature of weather prediction. In order to solve the problem, a nonlinear and symbolic regression method using GP(Genetic Programming) is suggested for a development of MOS for wind speed prediction. The proposed method is compared to various linear regression methods for prediction of wind speed. Also, statistical analysis of distribution for UM elements for each method is executed. experiments are performed for KLAPS(Korea Local Analysis and Prediction System) re-analysis data from 2007 to 2013 year for Jeju Island and Busan area in South Korea.

• Korean System Dynamics Review
• /
• v.9 no.1
• /
• pp.125-141
• /
• 2008

It is very difficult to predict future demand/supply, price for resources with acceptable accuracy using regression analysis. We try to use system dynamics to forecast the demand/supply and price for nickel. Nickel is very expensive mineral resource used for stainless production or other industrial production like battery, alloy making. Recent nickel price trend showed non-linear pattern and we anticipated the system dynamic method will catch this non-linear pattern better than the regression analysis. Our model has been calibrated for the past 6 year quarterly data (2002-2007) and tested for next 5 year quarterly data(2008-2012). The results were acceptable and showed higher accuracy than the results obtained from the regression analysis. And we ran the simulations for scenarios made by possible future changes in demand or supply related variables. This simulations implied some meaningful price change patterns.

• Earthquakes and Structures
• /
• v.15 no.4
• /
• pp.443-452
• /
• 2018

In this paper the new intensity measures (IMs) for probabilistic seismic analysis of RC high-rise buildings with core wall structural system are proposed. The existing IMs are analysed and the new optimal ones are presented. The newly proposed IMs are based on the existing ones which: 1) comprise a wider range of frequency velocity spectrum content and 2) are defined as the integral along the velocity spectrum. In analysis characteristics of optimal IMs such as: efficiency, practicality, proficiency and sufficiency are considered. As prototype buildings, RC high-rise buildings with core wall structural system and with characteristic heights: 20-storey, 30-storey and 40-storey, are selected. The non-linear 3D models of the prototype buildings are constructed. 720 non-linear time-history analyses are conducted for 60 ground motion records with a wide range of magnitudes, distances to source and various soil types. Statistical processing of results and detailed regression analysis are performed and appropriate demand models which relate IMs to demand measures (DMs), are obtained. The conducted analysis has shown that the newly proposed IMs can efficiently predict the DMs with minimum dispersion and satisfactory practicality as compared to the other commonly used IMs (e.g., PGA and $S_a(T_1)$). The newly proposed IMs overcome difficulties in calculating of integral along the velocity spectrum and present adequate replacement for IMs which comprise a wider range of frequency velocity spectrum content.

• Communications for Statistical Applications and Methods
• /
• v.20 no.2
• /
• pp.147-155
• /
• 2013

To determine whether each coefficient is equal to zero or not, usual $t$-tests are a popular choice (among others) in linear regression to practitioners because all statistical packages provide the statistics and their corresponding $p$-values. Under smaller samples (especially with non-normal errors) the tests often fail to correctly detect statistical significance. We propose a permutation approach by adopting a sufficient dimension reduction methodology to overcome this deficit. Numerical studies confirm that the proposed method has potential advantages over the t-tests. In addition, data analysis is also presented.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.371-374
• /
• 2008

In very high-spatial resolution remote sensing imagery, it is difficult to extract the feature information of various objects because of occlusion and shadows. Moreover, various and feeble information within shadows can be of use in GIS-based applications and remote sensing analysis. In this paper, we developed a radiometric restoration method for shadow areas using KOMPSAT-2 satellite image. After detecting the shadow, non-shadow pixels nearby are extracted using a morphological filter. An iterative linear regression method is applied to calculate the relationship between shadow and non-shadow pixels. The shadows are restored by the parameters of the linear regression algorithm. Tests show that recovery of shadowed areas by our method leads to improved image quality.