• Journal of the Korean Data and Information Science Society
• /
• v.18 no.3
• /
• pp.833-845
• /
• 2007

When X and Y have independent exponential distributions, we develop a Bayesian testing procedure for the ratio of two quantiles under reference prior. The noninformative prior such as reference prior is usually improper which yields a calibration problem that makes the Bayes factor to be defined up to a multiplicative constant. So we develop a Bayesian testing procedure based on fractional Bayes factor and intrinsic Bayes factor. We show that the posterior density under the reference prior is proper and propose the Bayesian testing procedure for the ratio of two quantiles using fractional Bayes factor and intrinsic Bayes factor. Simulation study and a real data example are provided.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.2
• /
• pp.153-165
• /
• 2003

This paper considers the likelihood ratio test statistic based on nonlinear regression quantiles estimators in order to test of hypothesis about the regression parameter $\theta_o$ and derives asymptotic distribution of proposed test statistic under the null hypothesis and a sequence of local alternative hypothesis. The paper also investigates asymptotic relative efficiency of the proposed test to the test based on the least squares estimators or the least absolute deviation estimators and gives some examples to illustrate the application of the main result.

• The Korean Journal of Applied Statistics
• /
• v.25 no.1
• /
• pp.101-114
• /
• 2012

In car insurance, the loss ratio is the ratio of total losses paid out in claims divided by the total earned premiums. In order to minimize the loss to the insurance company, estimating extreme quantiles of loss ratio distribution is necessary because the loss ratio has essential prot and loss information. Like other types of insurance related datasets, the distribution of the loss ratio has heavy-tailed distribution. The Peaks over Threshold(POT) and the Hill estimator are commonly used to estimate extreme quantiles for heavy-tailed distribution. This article compares and analyzes the performances of various kinds of parameter estimating methods by using a simulation and the real loss ratio of car insurance data. In addition, we estimate extreme quantiles using the Hill estimator. As a result, the simulation and the loss ratio data applications demonstrate that the POT method estimates quantiles more accurately than the Hill estimation method in most cases. Moreover, MLE, Zhang, NLS-2 methods show the best performances among the methods of the GPD parameters estimation.

• Communications of the Korean Mathematical Society
• /
• v.20 no.1
• /
• pp.145-159
• /
• 2005

In this paper we consider the problem of testing statistical hypotheses for unknown parameters in nonlinear regression models and propose three asymptotically equivalent tests based on regression quantiles estimators, which are Wald test, Lagrange Multiplier test and Likelihood Ratio test. We also derive the asymptotic distributions of the three test statistics both under the null hypotheses and under a sequence of local alternatives and verify that the asymptotic relative efficiency of the proposed test statistics with classical test based on least squares depends on the error distributions of the regression models. We give some examples to illustrate that the test based on the regression quantiles estimators performs better than the test based on the least squares estimators of the least absolute deviation estimators when the disturbance has asymmetric and heavy-tailed distribution.

• Journal of Korea Water Resources Association
• /
• v.41 no.4
• /
• pp.405-412
• /
• 2008

Rainfall quantiles were estimated by applying the FORGEX method. The circle network and two elliptical ones with the ratios of 1 to 1.5 and 1 to 2.0 were used and compared to find appropriate one for rainfall data. Annual maximum data were collected from 376 sites and standardized by the median. The networks were organized from the subject sites and then pooled and netmax data were collected from each network. Then, the growth curves and quantiles were estimated. When the subject site had small differences of quantiles from index flood method and at-site frequency analysis, those of the estimated quantiles from circle and elliptical networks were small. In contrast, the sites where the quantile differences are big have big differences of quantiles from circle and elliptical networks. The estimated quantiles from the elliptical network are more accurate than those from the circle network, because the ellipse network contains more sites in South Korea. Moreover, the ellipse with ratio of 1 to 2.0 shows closer quantiles to those from index flood method than one with ratio of 1 to 1.5. It is, therefore, found that the FORGEX method with 1 to 2.0 ellipse network is appropriate regional frequency analysis in South Korea.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.5
• /
• pp.61-72
• /
• 2020

The purpose of this study is to investigate the determinants of the capital structure of firms operating in a developing economy, Pakistan. The quantile regression method is applied on a sample of 183 non-financial companies listed on the Pakistan Stock Exchange during the period of 2008-2017. Specifically, the empirical analysis focuses on changes in the coefficients of the determinants according to the leverage ratio quantiles of the examined listed firms. The findings show that the capital structure of Pakistan listed firms differs between firms in different quantiles of leverage. These differences are significant with the sign of explanatory variables changes with the level of leverage. The research result found tangibility, profitability and age to be positively related to leverage among listed firms in Pakistan. However, size, liquidity and non-debt tax shield (NDTS) are negatively related to leverage. A firm's growth and risk are found to be insignificant predictors of capital structure in Pakistan listed firms. Moreover, the study also found a significant impact of industry characteristic on leverage. The findings of this study indicate that an individual firm's finance policy needs to be responsive to the firm's characteristics and should match with the different borrowing requirements of listed firms.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.1
• /
• pp.119-130
• /
• 2017

In many literatures on VaR and CTE for multivariate distribution, these are estimated by using transformed univariate distribution with a specific ratio of many kinds of portfolios. Even though there are lots of works to define quantiles for multivariate distributions, there does not exist a quantile uniquely. Hence, it is not easy to define the VaR and CTE. In this paper, we propose the weighted CTE vectors corresponding to various ratio combinations of many kinds of portfolios by extending the researches on the alternative VaR and integrated multivariate CTE based on multivariate quantiles. We extend relation equations about univariate CTEs to multivariate CTE vectors and discuss their characteristics. The proposed weighted CTEs are explored with some data from multivariate normal distribution and illustrative examples.

• Journal of Korean Society for Quality Management
• /
• v.49 no.2
• /
• pp.201-211
• /
• 2021

Purpose: This paper proposes a density adaptive grid algorithm for the k-NN regression model to reduce the computation time for large datasets without significant prediction accuracy loss. Methods: The proposed method utilizes the concept of the grid with centroid to reduce the number of reference data points so that the required computation time is much reduced. Since the grid generation process in this paper is based on quantiles of original variables, the proposed method can fully reflect the density information of the original reference data set. Results: Using five real-life datasets, the proposed k-NN regression model is compared with the original k-NN regression model. The results show that the proposed density adaptive grid-based k-NN regression model is superior to the original k-NN regression in terms of data reduction ratio and time efficiency ratio, and provides a similar prediction error if the appropriate number of grids is selected. Conclusion: The proposed density adaptive grid algorithm for the k-NN regression model is a simple and effective model which can help avoid a large loss of prediction accuracy with faster execution speed and fewer memory requirements during the testing phase.

• Journal of Korea Water Resources Association
• /
• v.38 no.1
• /
• pp.1-9
• /
• 2005

The study shows the possible use of the index flood frequency curves for an estimation of flood quantiles at ungauged locations. Flood frequency analysis were made for the annual maximum flood data series at 9 available stations in the Han river basin. From the flood frquency curve at each station the mean annual flood of 2.33-year return period was determined and the ratios of the flood magnitude of various return period to the mean annual flood at each station were averaged throughout the Han river basin, resulting mean flood ratios of different return periods. A correlation analysis was made between the mean annual flood and physiographic parameters of the watersheds i.e, the watershed area and mean river channel slope, resulting an empirical multiple linear regression equation over the whole Han river basin. For unguaged watershed the flood of a specified return period could be estimated by multiplying the mead flood ratio corresponding the return period with the mean annual flood computed by the empirical formula developed in terms of the watershed area and river channel slope. To verify the applicability of the methodology developed in the present study the floods of various return periods determined for the watershed in the river channel improvement plan formulation by the Ministry of Construction and Transportation(MOCT) were compared with those estimated by the present method. The result proved a resonable agreement up to the watershed area of approximately 2,000k $m^2$. It is suggested that the practice of design flood estimation based on the rainfall-runoff analysis might have to be reevaluated because it involves too much uncertainties in the hydrologic data and rainfall-runoff model calibration.

• Journal of Korea Water Resources Association
• /
• v.30 no.2
• /
• pp.177-191
• /
• 1997

The influences of the space allocation of design rainfall and partition of the subbasin on the characteristics of urban storm runoff was investigated for the 6 drainage basins by applying SWMM model. It show the deviation of -54.68∼18.77% in the peak discharge when we applied the composed JUFF quantiles to the two zones which are divided by upper and lower region of the basin. Then it is compared with the value for the case of using uniform rainfall distribution all over the drainage. Therefore, it would be helpful to decrease the flood risk when we adopt the space distribution of the design rainfall. The effects of the partitioning the drainage on the computing result shows various responses because of the surface characteristics of the each basin such as slope, imperviousness ratio, buy we can get closer result to the measured value as we make the subbasin detailed. If we use the concept of the skewness and area ratio when we determine the width of subbasin, we can improve the computed result even with fewer number of subbasins. We expect reasonable results which close into the measured results in the range of relative error, 25%, when we divide the basin into more than 3 subbasins and the total urban drainage area is less than 10$\textrm{km}^2$.