• Communications for Statistical Applications and Methods
• /
• v.5 no.3
• /
• pp.591-597
• /
• 1998

We prove (at least empirically) that some forms of the scaled residuals calculated from i.i.d. multivariate normal random vectors are ancillary. We further show that, if the scaled residuals are ancillary, then they have the same distribution whatever form of rotation is rosed to remove sample correlations.

• Communications for Statistical Applications and Methods
• /
• v.7 no.2
• /
• pp.385-392
• /
• 2000

Many tests for multivariate normality are based on the spherical coordinates of the scaled residuals of multivariate observations. Moore and Stubblebine's (1981) Pearson chi-square test is based on the radii of the scaled residuals, or equivalently the sample Mahalanobis distances of the observations from the sample mean vector. The chi-square statistic does not have a limiting chi-square distribution since the unknown parameters are estimated from ungrouped data. We will derive a simple closed form of the Rao-Robson chi-square test statistic and provide a self-contained proof that it has a limiting chi-square distribution. We then provide an illustrative example of application to a real data with a simulation study to show the accuracy in finite sample of the limiting distribution.

• Communications for Statistical Applications and Methods
• /
• v.22 no.4
• /
• pp.361-375
• /
• 2015

A measure of skewness and kurtosis is proposed to test multivariate normality. It is based on an empirical standardization using the scaled residuals of the observations. First, we consider the statistics that take the skewness or the kurtosis for each coordinate of the scaled residuals. The null distributions of the statistics converge very slowly to the asymptotic distributions; therefore, we apply a transformation of the skewness or the kurtosis to univariate normality for each coordinate. Size and power are investigated through simulation; consequently, the null distributions of the statistics from the transformed ones are quite well approximated to asymptotic distributions. A simulation study also shows that the combined statistics of skewness and kurtosis have moderate sensitivity of all alternatives under study, and they might be candidates for an omnibus test.

• Communications for Statistical Applications and Methods
• /
• v.30 no.4
• /
• pp.423-435
• /
• 2023

The Jarque and Bera (1980) statistic is one of the well known statistics to test univariate normality. It is based on the sample skewness and kurtosis which are the sample standardized third and fourth moments. Desgagné and de Micheaux (2018) proposed an alternative form of the Jarque-Bera statistic based on the sample second power skewness and kurtosis. In this paper, we generalize the statistic to a multivariate version by considering some data driven directions. They are directions given by the normalized standardized scaled residuals. The statistic is a modified multivariate version of Kim (2021), where the statistic is generalized using an empirical standardization of the scaled residuals of data. A simulation study reveals that the proposed statistic shows better power when the dimension of data is big.

• Journal of Gastric Cancer
• /
• v.15 no.4
• /
• pp.238-245
• /
• 2015

Purpose: This study aimed to identify time-dependent prognostic factors and demonstrate the time-dependent effects of important prognostic factors in patients with advanced gastric cancer (AGC). Materials and Methods: We retrospectively evaluated 3,653 patients with AGC who underwent curative standard gastrectomy between 1991 and 2005 at the Korea Cancer Center Hospital. Multivariate survival analysis with Cox proportional hazards regression was used in the analysis. A non-proportionality test based on the Schoenfeld residuals (also known as partial residuals) was performed, and scaled Schoenfeld residuals were plotted over time for each covariate. Results: The multivariate analysis revealed that sex, depth of invasion, metastatic lymph node (LN) ratio, tumor size, and chemotherapy were time-dependent covariates violating the proportional hazards assumption. The prognostic effects (i.e., log of hazard ratio [LHR]) of the time-dependent covariates changed over time during follow-up, and the effects generally diminished with low slope (e.g., depth of invasion and tumor size), with gentle slope (e.g., metastatic LN ratio), or with steep slope (e.g., chemotherapy). Meanwhile, the LHR functions of some covariates (e.g., sex) crossed the zero reference line from positive (i.e., bad prognosis) to negative (i.e., good prognosis). Conclusions: The time-dependent effects of the prognostic factors of AGC are clearly demonstrated in this study. We can suggest that time-dependent effects are not an uncommon phenomenon among prognostic factors of AGC.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.4 no.3
• /
• pp.697-703
• /
• 2000

As device dimensions are lastly scaled down, impact ionization(I.I.) events are very important to analyze hot carrier transport in high energy region, and the exact model of impact ionization is demanded on device simulation. We calculate full band model by empirical pseudopotential method and the impact ionization rate is derived from modified Keldysh formula. We calculate impact ionization coefficients by full band Monte Carlo simulator to investigate temperature dependent characteristics of impact ionization for GaAs as a function of field. Resultly impact ionization coefficients are in good agreement with experimental values at look. We how energy is increasing along increasing the field, while energy is decreasing along increasing the temperature since the phonon scattering rates for emission mode are very high at high temperature. The logarithmic fitting function of impact ionization coefficients is described as a second orders function of temperature and field. The residuals of the logarithmic fitting function are mostly within 5%. We Dow, therefore, the logarithm of impact ionization coefficients has quadratic dependence on temperature, and we can save time of calculating the temperature dependent impact ionization coefncients as a function of field.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2000.05a
• /
• pp.460-464
• /
• 2000

As device dimensions are lastly scaled down, impact ionization(I.I.) events are very important to analyze hot carrier transport in high energy region, and the exact model of impact ionization is demanded on device simulation. We calculate full band model by empirical pseudopotential method and the impact ionization rate is derived from modified Keldysh formula. We calculate impact ionization coefficients by full band Monte Carlo simulator to investigate temperature-and field-dependent characteristics of impact ionization for GaAs. Resultly impact ionization coefficients are In good agreement with experimental values at 300k. We know energy is increasing along increasing the field. while energy is decreasing along increasing the temperature since the phonon scattering rates for omission mode are very high at high temperature. The logarithmic fitting function of impact ionization coefficients is described as a second orders function for temperature and field. The residuals of the logarithmic fitting function are mostly within 5%. We know, therefore, logarithm of impact ionization coefficients has quadratic dependence on temperature and field, and we can save time of calculating the temperature- and field-dependent impact ionization coefficients.