• Communications for Statistical Applications and Methods
• /
• v.15 no.6
• /
• pp.825-835
• /
• 2008

Multivariate GARCH(MGARCH) has been useful in financial studies and econometrics for modeling volatilities and correlations between components of multivariate time series. An obvious drawback lies in that the number of parameters increases rapidly with the number of variables involved. This thesis tries to resolve the problem by using dimension reduction technique. We briefly review both factor models for dimension reduction and the MGARCH models including EWMA (Exponentially weighted moving-average model), DVEC(Diagonal VEC model), BEKK and CCC(Constant conditional correlation model). We create meaningful portfolios obtained after reducing dimension through statistical factor models and fundamental factor models and in turn these portfolios are applied to MGARCH. In addition, we compare portfolios by assessing MSE, MAD(Mean absolute deviation) and VaR(Value at Risk). Various financial time series are analyzed for illustration.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1285-1285
• /
• 2001

For NIR measurements of papers normally diffuse reflectance accessories are used which can provide a large sampling area. The in-line process control FT-NIR spectrometer MATRIX-E enables the contactless measurement of paper samples of low silicone coat weights on label-stocks in a paper converting factory. For this study concentrations of silicone between 0 and 2 g/$m^2$ on various paper substrates were included in a quantitative method. The aim was to achieve an absolute value for the deviation from the target value of 1 g/$m^2$ during continuous movement of the paper with velocities around 400 numinute. Influences from the uncoated paper type due to supplier, color, opacity, area densities, pre-coating as well as different compounds of the agent silicone were investigated and it was found that all these papers can be represented in one PLS-model. Especially the fact that silicone as an element is present in clay coated papers is of no consequence to the measurements with MATRIX-E. Moreover during in-line installations the variation of the moisture contents in the moving paper due to variable machine velocities as well as the reflecting material of the cylinder have to be considered. It is shown that the result of the in-line calibration has the same prediction ability compared to lab scale results(Root Mean Square Error of Cross-Validation RMSECV = 0.034 g/$m^2$).

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1193-1193
• /
• 2001

For NIR measurements of papers normally diffuse reflectance accessories are used which can provide a large sampling area. The in-line process control FT-NIR spectrometer MATRIX-E enables the contactless measurement of paper samples of low silicone coat weights on label-stocks in a paper converting factory. For this study concentrations of silicone between 0 and 2 g/㎡ on various paper substrates were included in a quantitative method. The aim was to achieve an absolute value for the deviation from the target value of 1 g/㎡ during continuous movement of the paper with velocities around 400 m/minute. Influences from the uncoated paper type due to supplier, color, opacity, area densities, pre-coating as well as different compounds of the agent silicone were investigated and it was found that all these papers can be represented in one PLS-model. Especially the fact that silicone as an element is present in clay coated papers is of no consequence to the measurements with MATRIX-E. Moreover during in-line installations the variation of the moisture contents in the moving paper due to variable machine velocities as well as the reflecting material of the cylinder have to be considered. It is shown that the result of the in-line calibration has the same prediction ability compared to lab scale results (Root Mean Square Error of Cross-Validation RMSECV = 0.034 g/㎡).

• Journal of the Korean Society of Safety
• /
• v.21 no.3 s.75
• /
• pp.45-52
• /
• 2006

The AITs(autoignition temperatures) describe the minimum temperature to which a substance must be heated, without the application of a flame or spark, which will cause that substance to ignite. The AITs are often used as a factor in determining the upper temperature limit for processing operations and conditions for handling, storage and transportation, and in determining potential fire hazard from accidental contact with hot surfaces. The measurement AITs are dependent upon many factors, namely initial temperature, pressure, volume, fuel/air stoichiometry, catalyst material, concentration of vapor, time lag. Therefore, the AITs reported by different ignition conditions are sometimes significantly different. This study measured the AITs of benzene, toluene and xylene isomers from time lag using AS1M E659-78 apparatus. The experimental ignition delay times were a good agreement with the calculated ignition delay times by the proposed equations wtih a few A.A.D.(average absolute deviation). Also The experimental AITs of benzene, toluene, o-xylene, m-xylene and p-xylene were $583^{\circ}C,\;547^{\circ}C,\;480^{\circ}C,\;587^{\circ}C,\;and\;557^{\circ}C$, respectively.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2000.04a
• /
• pp.91-98
• /
• 2000

Derivation of stage-discharge relationship and characteristics for Yangsan river is presented in this paper. This research has been conducted as the second one after the first trial of 1997. The determination of discharge at a Yangsan river gauging was best made by measuring the flow velocities with a current meter and rod float. The rating curve obtained through 52 stage-discharge measurements on Yangsan river basin in 1999 is represented by Q=15.3540-140.6076H+182.44372$H^2$, which is discovered to be most excellent among other curves in reliability analysis. The capability of the observed stage-discharge data for Yangsan river was tested by HEC-RAS program, and its capability to reproduce discharge was investigated and compared with the computational results. Rating curve stability is determined on the basis of deviations in the stage-discharge relationship, utilization of specific gauge, and absolute differences between sequential stream flow measurements and an analysis residuals. Therefore it seems necessary to research method to obtain rating curve in a rigorous and accurate manner.

• Atmosphere
• /
• v.21 no.4
• /
• pp.361-371
• /
• 2011

In this study, characteristics and trends of tropical night (TN) are investigated by using the KMA 14 observation data for the recent 50 years (1958-2007) over South Korea. The TN is defined as a day with a daily minimum temperature exceeds the absolute threshold temperature ($25^{\circ}C$), and the relative deviation from normal temperature, 95th percentile of all observed daily minimum temperature. Although the spatial distribution of TN occurrence depends on the choice of the definitions, the frequency of TN shows strong spatial and interannual variations with the minimum at high land area (Chupungnyeong and wet years) and maximum at southern coastal area and large city area (Jeju, Busan, Seoul, Daegu). Most of TN occurs in August (56%) and July (41%), and the duration date of TN is proportional to the frequency of TN without regard to the definition method. In general, increasing trends are found in the TN time series without regard to the analysis method, but the trends are clearly depending on the analysis period and geographic locations. Decreasing trends are prominent during the most of analysis period, especially until the mid of 1990, whereas strong increasing trends are found during recent 30 years (1988-2007), especially at Jeju, Ulsan, Daegu and Pohang. Also the severity of TN is significantly increased in recent years.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.19 no.6
• /
• pp.457-463
• /
• 2007

Various correlations of saturated vapor density in a truncated power series form are tested and compared in this study. Saturated vapor density correlation can be expressed relating logarithmic reduced density to the reduced temperature. Five types of correlation has been investigated using saturated vapor density data for 22 pure substance refrigerants from ASHRAE (American Society of Heating, Reftigerating and Air-Conditioning Engineers, Inc.) property tables and NIST (National Institute of Standards and Technology) Chemistry Webbook. Correlations are fitted to the data points by least squares method. Data points are equally weighted. The best type of correlation among the five types is suggested. The results obtained indicate that the best correlations with 3, 4, and 5 terms yield average AAD's (Average Absolute Deviation) of 0.27%, 0.04%, and 0.02%, respectively, while widely used conventional correlations with 3, 4, and 5 terms yield those of 1.19%, 0.61%, and 0.17%. The suggested type of correlation could reduce the number of terms while improving performance.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.2
• /
• pp.61.1-61.1
• /
• 2016

Black hole (BH) mass is a fundamental quantity to understand BH growth, galaxy evolution, and connection between them. Thus, obtaining accurate and precise BH mass estimates over cosmic time is of paramount importance. The rest-frame UV CIV ${\lambda}1549$ broad emission line is commonly used for BH mass estimates in high-redshift AGNs (i.e., $2{\leq}z{\leq}5$) when single-epoch (SE) optical spectra are available. Achieving correct and accurate calibration for CIV-based SE BH mass estimators against the most reliable reverberation-mapping based BH mass estimates is thus practically important and still useful. By performing multi-component spectral decomposition analysis to obtained high-quality HST UV spectra for the updated sample of local reverberation-mapped AGNs including new HST STIS observations, CIV emission line widths and continuum luminosities are consistently measured. Using a Bayesian hierarchical model with MCMC sampling based on Hamiltonian Monte Carlo algorithm (Stan NUTS), we provide the most consistent and accurate calibration of CIV-based BH mass estimators for the three line width characterizations, i.e., full width at half maximum (FWHM), line dispersion (${\sigma}_{line}$), and mean absolute deviation (MAD), in the extended BH mass dynamic range of log $M_{BH}/M_{\odot}=6.5-9.1$.

• Steel and Composite Structures
• /
• v.21 no.4
• /
• pp.921-947
• /
• 2016

Concrete Filled Fibre Reinforced Polymer Tube (CFFT) for new columns construction has attracted significant research attention in recent years. The CFFT acts as a formwork for new columns and a barrier to corrosion accelerating agents. It significantly increases both the strength capacity (Strength enhancement ratio) and the ductility (Strain enhancement ratio) of reinforced concrete columns. In this study, based on predefined selection criteria, experimental investigation results of 134 circular CFFT columns under axial compression have been compiled and analysed from 599 CFFT specimens available in the literature. It has been observed that actual confinement ratio (expressed as a function of material properties of fibres, diameter of CFFT and compressive strength of concrete) has significant influence on the strength and ductility of circular CFFT columns. Design oriented models have been proposed to compute the strength and strain enhancement ratios of circular CFFT columns. The proposed strength and strain enhancement ratio models have significantly reduced Average Absolute Error (AAE), Mean Square Error (MSE), Relative Standard Error of Estimate (RSEE) and Standard Deviation (SD) as compared to other available strength and strain enhancement ratios of circular CFFT column models. The predictions of the proposed strength and strain enhancement ratio models match well with the experimental strength and strain enhancement ratios investigation results in the compiled database.

• Proceedings of the SAREK Conference
• /
• 2009.06a
• /
• pp.804-809
• /
• 2009

The present study measured frost pattern on a cylinder to propose empirical correlation equations for the local and average frost thickness, frost density and frost mass. The key parameters were diameter of the cylinders (7mm, 20mm), cooling surface temperature of the circular tube, absolute humidity of air, air temperature and air velocity. A 50% ethylene glycol aqueous solution was applied as a coolant. The frost thicknesses at both front and rear were larger than those at the other parts, while they were increased as diameter of the cylinder was increased. The local frost thicknesses at high air velocity were more uniform than those at low air velocity. The values predicted by Kim et al. under the freezer condition showed larger by the maximum of $30{\sim}50%$ than the measured data under heat pump condition. The empirical correlations for the local and average frost thickness and frost mass were proposed. The correlation equations for the frost thickness and frost mass under the heat pump condition in the present study might predict more accurate than the other correlation equations. The proposed correlations might be applied for the freezer condition within the maximum 15% deviation from the previous correlations under freezer condition.