• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• pp.853-864
• /
• 2002

In the analysis of categorical data subject to misclassification errors, the observed cell proportions are adjusted by a misclassification probabilities and estimates of variances are adjusted accordingly. In this case, it is important to determine the extent to which misclassification probabilities are homogeneous within a population. This paper considers methods to evaluate the power of chi-squared tests for homogeneity with complex survey data subject to misclassification errors. Two cases are considered: adjustment with homogeneous misclassification probabilities; adjustment with heterogeneous misclassification probabilities. To estimate misclassification probabilities, logistic regression method is considered.

• Journal of the Korean Statistical Society
• /
• v.36 no.2
• /
• pp.275-284
• /
• 2007

Confidence intervals for the population proportion in a double sampling scheme subject to false-positive misclassification are considered. The confidence intervals are obtained by applying Agresti and Coull's approach, so-called "adding two-failures and two successes". They are compared in terms of coverage probabilities and expected widths with the Wald interval and the confidence interval given by Boese et al. (2006). The latter one is a test-based confidence interval and is known to have good properties. It is shown that the Agresti and Coull's approach provides a relatively simple but effective confidence interval.

• Journal of the Korean Statistical Society
• /
• v.27 no.4
• /
• pp.471-483
• /
• 1998

The influence of observations on the misclassification probability in multiple discriminant analysis under the equal covariance assumption is investigated by the local influence method. Under an appropriate perturbation we can get information about influential observations and outliers by studying the curvatures and the associated direction vectors of the perturbation-formed surface of the misclassification probability. We show that the influence function method gives essentially the same information as the direction vector of the maximum slope. An illustrative example is given for the effectiveness of the local influence method.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.1
• /
• pp.217-223
• /
• 2016

Outcome misclassification is widespread in classification problems, but methods to account for it are rarely used. In this paper, the problem of inference with misclassified multinomial logit data with a large number of multinomial parameters is addressed. We have had a significant swell of interest in the development of novel methods to infer misclassified data. One simulation study is shown regarding how seriously misclassification issue occurs if the number of categories increase. Then, using the group lasso regression, we will show how the best model should be fitted for that kind of multinomial regression problems comprehensively.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.11a
• /
• pp.365-368
• /
• 2006

The implementation of a screening procedure for removing non-conforming products has become a common practice especially in high-tech manufacturing industries. Screening procedures involve a measurement on the quality characteristic of interest since decisions regarding the conformance to specifications are usually made on the basis of the realization of measurement. A significant variability in measurement procedures may result in the misclassification of an outgoing product (that is, falsely accepting defectives or falsely rejecting conforming items), which may lead to wrong interpretation on product quality, It may thus be necessary to consider the impacts of misclassification errors due to measurement variability when designing screening procedures. Along this line, this article investigates the design of screening procedures based on the assessment of misclassification errors. The main objective is to determine the screening limits on measured values so that two types of misclassification errors may properly be compromised.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.3
• /
• pp.522-532
• /
• 1995

Multilayer neural networks with backpropagation learning algorithm are widely used for pattern classification problems. For many real applications, it is more important to reduce the misclassification rate than to increase the rate of successful classification. But multilayer perceptrons(MLP's) have drawbacks of slow learning speed and false convergence to local minima. In this paper, we propose a new method for character recognition problems with a single-layer network and double rejection mechanisms, which guarantees a very low misclassification rate. Comparing to the MLP's, it yields fast learning and requires a simple hardware architecture. We also introduce a new coding scheme to reduce the misclassification rate. We have prepared two databases: one with 135,000 digit patterns and the other with 117,000 letter patterns, and have applied the proposed method for printed character recognition, which shows that the method reduces the misclassification rate significantly without sacrificing the correct recognition rate.

• Journal of the Korean Statistical Society
• /
• v.25 no.1
• /
• pp.145-151
• /
• 1996

The local behaviour of the surface formed by the perturbed maximum likelihood estimator of the squared Mahalanobis distance is investigated. The study of the local behaviour allows a simultaneous perturbation on the samples of interest and it is effective in identifying influential observations.

• The Korean Journal of Applied Statistics
• /
• v.14 no.1
• /
• pp.39-57
• /
• 2001

Categorical data has sometimes misclassification errors. If this data will be analyzed, then estimated cell probabilities could be biased and the standard Pearson X2 tests may have inflated true type I error rates. On the other hand, if we regard wellclassified data with misclassified one, then we might spend lots of cost and time on adjustment of misclassification. It is a necessary and important step to ask whether categorical data is misclassified before analyzing data. In this paper, when data is misclassified at one of two variables for two-dimensional contingency table and marginal sums of a well-classified variable are fixed. We explore to partition marginal sums into each cells via the concepts of Bound and Collapse of Sebastiani and Ramoni (1997). The double sampling scheme (Tenenbein 1970) is used to obtain informations of misclassification. We propose test statistics in order to solve misclassification problems and examine behaviors of the statistics by simulation studies.

• The Korean Journal of Archival Studies
• /
• no.58
• /
• pp.299-332
• /
• 2018

The purpose of this study is to investigate the current status and causes of misclassification of school records and to suggest the directions to improve the School Records Management Criteria Table(general disposition authority for school records), which will lead to misclassification reducement. This study begins with analysing the records created or received in four schools sampled for one year to investigate the status and causes of misclassifications. A advisory group including four administrative officers and seven records managers was formed and group meeting was held twice to identify the causes of the misclassification and to suggest alternatives. In this study, 33 unit tasks(transactions) with frequent misclassification were identified, and the cause of misclassification was analyzed based on focus group interviews. The main causes of misclassification were categorized into two types. This study concludes with suggesting the improvement of the School Records Management Criteria Table for addressing the causes, including commentary reinforcement and the addition of workflow to complex tasks.

• Clinical and Experimental Pediatrics
• /
• v.64 no.3
• /
• pp.96-102
• /
• 2021

This study aimed to investigate the effect of maternal smoking during pregnancy (MSDP) on the risk of bone fractures in the offspring through a systematic review and meta-analysis. The PubMed, Web of Science, and Scopus databases were systematically searched for relevant articles published through July 2019. According to heterogeneity, the pooled risk ratio (RR) and odds ratio (OR) and their corresponding 95% confidence interval (CI) were obtained using fixed or random effects models. The heterogeneity and quality of the included studies were assessed by the I-squared (I²) statistic and the Newcastle-Ottawa scale, respectively. Sensitivity analyses were performed to test the effect of MSDP misclassification on the results. The review of 842 search records yielded 5 studies including 8,746 mother-child pairs that were included in the meta-analysis. Pooling adjusted effect measures showed that MSDP was not associated with a later risk of bone fractures in the offspring (pooled RR, 1.15; 95% CI, 0.84-1.58; I²=66.8%; P=0.049). After the adjustment for misclassification, MSDP may be associated with a 27% increased risk of bone fracture (pooled OR, 1.27; 95% CI, 1.00-1.62; I²=0%; P=0.537). After the adjustment for misclassification, MSDP is associated with an increased risk of bone fractures among children whose mothers smoked during pregnancy.