• Communications for Statistical Applications and Methods
• /
• 제26권2호
• /
• pp.205-216
• /
• 2019

Diagnostic tests in medical fields detect or diagnose a disease with results measured by continuous or discrete ordinal data. The performance of a diagnostic test is summarized using the receiver operating characteristic (ROC) curve and the area under the curve (AUC). The diagnostic test is considered clinically useful if the outcomes in actually-positive cases are higher than actually-negative cases and the ROC curve is concave. In this study, we apply the stochastic ordering method in a Bayesian hierarchical model to estimate the proper ROC curve and AUC when the diagnostic test results are measured in discrete ordinal data. We compare the conventional binormal model and binormal model under stochastic ordering. The simulation results and real data analysis for breast cancer indicate that the binormal model under stochastic ordering can be used to estimate the proper ROC curve with a small bias even though the sample sizes were small or the sample size of actually-negative cases varied from actually-positive cases. Therefore, it is appropriate to consider the binormal model under stochastic ordering in the presence of large differences for a sample size between actually-negative and actually-positive groups.

• Journal of the Korean Data and Information Science Society
• /
• 제4권
• /
• pp.1-11
• /
• 1993

The Cox's regression model is frequently used for covariate effects in survival data analysis, But, much of the statistical work has focused on asymptotic behavior so the small sample evaluation has been neglected. In this paper, we compare the small or moderate sample performances of the survival function estimators for the Cox's regression model using bootstrap method. The smoothed PL type estimator and the Link estimator are slightly better than corresponding the PL type estimator and the Nelson type estimator in the sense of the achieved error rates.

• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.553-566
• /
• 2003

In many cases, the measurement error variances may be functions of the unknown true values or related covariates. This paper considers design-based estimators of the parameters of these variance functions based on the within-unit sample variances. This paper devotes to: (1) define an error scale factor $\delta$; (2) develop estimators of the parameters of the linear measurement error variance function of the true values under large-sample and small-error conditions; (3) use propensity methods to adjust survey weights to account for possible selection effects at the replicate level. The proposed methods are applied to medical examination data from the U.S. Third National Health and Nutrition Examination Survey (NHANES III).

• Communications for Statistical Applications and Methods
• /
• 제24권3호
• /
• pp.291-301
• /
• 2017

It is well known in a small sample that the maximum likelihood (ML) approach for variance components in the general linear model yields estimates that are biased downward. The ML estimate of residual variance tends to be downwardly biased. The underestimation of residual variance, which has implications for the estimation of marginal effects and asymptotic standard error of estimates, seems to be more serious in some limited dependent variable models, as shown by some researchers. An alternative frequentist's approach may be restricted or residual maximum likelihood (REML), which accounts for the loss in degrees of freedom and gives an unbiased estimate of residual variance. In this situation, the REML estimator is derived in a censored regression model. A small sample the REML is shown to provide proper inference on regression coefficients.

• 응용통계연구
• /
• 제24권1호
• /
• pp.137-143
• /
• 2011

The classification of subjects with unknown distribution in a small sample size often involves order-restricted constraints in multivariate parameter setups. Those problems make the optimality of a conventional likelihood ratio based statistical inferences not feasible. Fortunately, Roy (1953) introduced union-intersection principle(UIP) which provides an alternative avenue. Multivariate linear rank statistics along with that principle, yield a considerably appropriate robust testing procedure. Furthermore, conditionally distribution-free test based upon exact permutation theory is used to generate p-values, even in a small sample. Applications of this method are illustrated in a real microarray data example (Lobenhofer et al., 2002).

• Journal of Astronomy and Space Sciences
• /
• 제11권2호
• /
• pp.154-164
• /
• 1994

The secular period behavior of 23 field RR Lyrae stars is studied in order to determine if the observed period changes could be attributed, at least in the mean, to stellar evolution. The sample of stars is subdivided into two Oosterhoff groups based on the metallicity and period-shift. Despite the small sample size, we found a distinct bias toward positive period changes in the group II variables. The period changes of the group I variables, however, are small and in the mean near zero. This is consistent with the behavior predicted by the recent evolutionary models, as was the case for the variables in globular clusters. This provides yet another support for the Lee, Demarque, and Zinn (1990) evolutionary models of RR Lyrase stars and their explanation of the Sandage period-shift effect.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 A
• /
• pp.255-258
• /
• 2000

The supplementalγ controls of the FACTS are designed for the enhancement of the small signal stability in power system. The designed supplementary controllers using residue are applied to SVC or TCSC for the improving the damping ratio of dominant eigen value in the New England and 39 bus test system as the sample system. The results show the validation of the supplementary controller for the enhancement of the eigenvalues which have the low frequency oscillations with poor damping ratio as the unstable problem in the sample system.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.109.3-109
• /
• 2001

The so-called curse of dimensionality arises when Gaussian mixture is used on high-dimensional small-sample-size data, since the number of free elements that needs to be specied in each covariance matrix of Gaussian mixture increases exponentially with the number of dimension d. In this paper, by constraining the covariance matrix in its decomposed orthonormal form we get a local PCA model so as to reduce the number of free elements needed to be specified. Moreover, to cope with the small sample size problem, we adopt BYY data smoothing learning which is a regularization over maximum likelihood learning obtained from BYY harmony learning to implement this local PCA model.

• 품질경영학회지
• /
• 제25권1호
• /
• pp.69-79
• /
• 1997

We consider the semiparametric linear errors-in-variables model: yi=(${\alpha}+{\beta}ui+{\varepsilon}i$, xi=ui＋${\varepsilon}i$ i=1, …, n where (xi, yi) stands for an observation vector, (ui) denotes a set of incidental nuisance parameters, (${\alpha}$ , ${\beta}$) is a vector of regression parameters and (${\varepsilon}i$, ${\delta}i$) are mutually uncorrelated measurement errors with zero mean and finite variances but otherwise unknown distributions. On the basis of a simple small-sample low-noise a, pp.oximation, we propose a new method of comparing the mean squared errors(MSE) of the various competing estimators of the true regression parameters ((${\alpha}$ , ${\beta}$). Then we show that a class of estimators including the classical least squares estimator and the maximum likelihood estimator are consistent and first-order efficient within the class of all regular consistent estimators irrespective of type of measurement errors.

• 대한산업공학회지
• /
• 제23권1호
• /
• pp.177-196
• /
• 1997

In this paper, statistically optimal accelerated life test(ALT) plans considering statistical efficiency only and new compromise ALT plans to sacrifice some statistical efficiency in return for improved overall properties including estimobility probability and robustness for the model assumptions are developed under the assumptions of constant stress, intermittent inspection, Type I censoring and lognormal failure distribution which has been one of the popular choices of failure distributions in the extensive engineering applications of ALT. Computational experiments are conducted to compare with four ALT plans including two proposed ones under continuous and intermittent inspections over a range of parameter values in terms of asymptotic variance, sensitivities for guessed input values, and proportion of estimable samples, etc. The small and moderate sample properties for the proposed ALT plans designed under asymptotic criterion are also investigated by Monte Carlo simulation.