• The Korean Journal of Applied Statistics
• /
• v.25 no.2
• /
• pp.341-350
• /
• 2012

Composite quantile regression model is considered for iid error case. Since the regression coefficients are the same across different quantiles, composite quantile regression can be used to combine the strength across multiple quantile regression models. For the composite quantile regression, bootstrap method is examined for statistical inference including the selection of the number of quantiles and confidence intervals for the regression coefficients. Feasibility of the bootstrap method is demonstrated through a simulation study.

• Communications for Statistical Applications and Methods
• /
• v.20 no.4
• /
• pp.259-270
• /
• 2013

This paper considers a penalized composite quantile regression (CQR) that performs a variable selection in the linear model with grouped variables. An adaptive sup-norm penalized CQR (ASCQR) is proposed to select variables in a grouped manner; in addition, the consistency and oracle property of the resulting estimator are also derived under some regularity conditions. To improve the efficiency of estimation and variable selection, this paper suggests the two-stage penalized CQR (TSCQR), which uses the ASCQR to select relevant groups in the first stage and the adaptive lasso penalized CQR to select important variables in the second stage. Simulation studies are conducted to illustrate the finite sample performance of the proposed methods.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.733-745
• /
• 2017

Local composite quantile regression is a useful non-parametric regression method widely used for its high efficiency. Data smoothing methods using kernel are typically used in the estimation process with performances that rely largely on the smoothing parameter rather than the kernel. However, $L_2$-norm is generally used as criterion to estimate the performance of the regression function. In addition, many studies have been conducted on the selection of smoothing parameters that minimize mean square error (MSE) or mean integrated square error (MISE). In this paper, we explored the optimality of selecting smoothing parameters that determine the performance of non-parametric regression models using local linear composite quantile regression. As evaluation criteria for the choice of smoothing parameter, we used mean absolute error (MAE) and mean integrated absolute error (MIAE), which have not been researched extensively due to mathematical difficulties. We proved the uniqueness of the optimal smoothing parameter based on MAE and MIAE. Furthermore, we compared the optimal smoothing parameter based on the proposed criteria (MAE and MIAE) with existing criteria (MSE and MISE). In this process, the properties of the proposed method were investigated through simulation studies in various situations.

• Atmosphere
• /
• v.28 no.4
• /
• pp.383-392
• /
• 2018

This study investigates synoptic characteristics of cold surges over South Korea during winter season (December-February). A total of 63 cold events are selected by quantile regression analysis using daily mean temperature observations from 11 KMA stations for 38 years (1979/80-2016/17). Large-scale pressure pattern during the cold surges is well characterized by high over Siberia and low over Aleutian regions, which elucidates cold advection over the Korean peninsula. However, the large-scale pattern cannot successfully explain the observed sudden decrease of temperature during the cold surges. Composite analyses reveal that a synoptic-scale cyclone developing over the northern Japan is a key feature that significantly contribute to the enhancement of cold advection by increasing pressure gradient over the Korean peninsula. Enhanced sensible and latent heat fluxes are observed over the southern ocean of Korea and Japan during the cold surges due to temperature and humidity differences between the near surface and the lower atmosphere over the ocean. The evaporated water vapor transported toward the center of the surface cyclone and condenses in the lower-to-middle troposphere. The released energy likely promotes the development of the surface cyclone by inducing positive PV near the surface of the heating region.

• Atmosphere
• /
• v.25 no.3
• /
• pp.435-447
• /
• 2015

This study explores the synoptic characteristics of cold days over South Korea and their relationship with large-scale climate variability. The cold day, which is different from cold surge, is defined when daily-mean surface air temperature, averaged over 11 KMA stations, is colder than 1-percentile temperature in each year by considering its long-term trend over 1960~2012. Such event is detected by quantile regression and the related synoptic patterns are identified in reanalysis data. Composite geopotential height anomalies at 500 hPa show that cold days are often preceded by positive anomalies in high latitudes and negative anomalies in midlatitudes on the west of Korea. While the formers are quasi-stationary and quasi-barotropic, and often qualified as blocking highs, the latters are associated with transient cyclones. At cold days, the north-south dipole in geopotential height anomalies becomes west-east dipole in the lower troposphere as high-latitude anticyclone expands equatorward to the Northern China and mid-latitude cyclone moves eastward and rapidly develops over the East Sea. The resulting northerlies cause cold days in Korea. By performing composite analyses of large-scale climate indices, it is further found that the occurrence of these cold days are preferable when the Arctic Oscillation is in its negative phase and/or East Asian monsoon circulation and Siberian high are anomalously strong.