• Proceedings of the Korean Statistical Society Conference
• /
• 2000.11a
• /
• pp.65-73
• /
• 2000

Kernel estimator is very popular in nonparametric density estimation. In this paper we propose an estimator which reduces the bias to the fourth power of the bandwidth, while the variance of the estimator increases only by at most moderate constant factor. The estimator is fully nonparametric in the sense of convex combination of three kernel estimators, and has good numerical properties.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.1
• /
• pp.213-220
• /
• 2006

Since Beran (1977) developed the minimum Hellinger distance estimation, this method has been a popular topic in the field of robust estimation. In the process of defining a distance, a kernel density estimator has been widely used as a density estimator. In this article, however, we show that a combination of a kernel density estimator and an empirical density could result a smaller bias of the minimum Hellinger distance estimator than using just a kernel density estimator for a location parameter.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.148-148
• /
• 2022

Spatiotemporal precipitation data is one of the primary quantities in hydrological as well as climatological studies. Despite the fact that the estimation of these data has made considerable progress owing to advances in remote sensing, the discrepancy between satellite-derived precipitation product (SPP) data and observed data is still remarkable. This study aims to propose an effective deep learning model (DLM) for bias correction of SPPs. In which TRMM (The Tropical Rainfall Measuring Mission), CMORPH (CPC Morphing technique), and PERSIANN-CDR (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) are three SPPs with a spatial resolution of 0.25o exploited for bias correction, and APHRODITE (Asian Precipitation - Highly-Resolved Observational Data Integration Towards Evaluation) data is used as a benchmark to evaluate the effectiveness of DLM. We selected the Mekong River Basin as a case study area because it is one of the largest watersheds in the world and spans many countries. The adjusted dataset has demonstrated an impressive performance of DLM in bias correction of SPPs in terms of both spatial and temporal evaluation. The findings of this study indicate that DLM can generate reliable estimates for the gridded satellite-based precipitation bias correction.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.13 no.6
• /
• pp.982-989
• /
• 2010

This paper shows accuracy comparison results of reliability estimation methods for one-shot systems such as ammunitions. Quantal-response data, following a binomial distribution at each sampling time, characterizes lifetimes of one-shot systems. Various quantal-response data of different sample sizes are simulated using lifetime data randomly sampled from assumed weibull distributions with different shape parameters but the identical scale parameter in this paper. Then, reliability estimation methods in open literature are applied to the simulated quantal-response data to estimate true reliability over time. Rankings in estimation accuracy for different sample sizes are determined using t-test of SSE. Furthermore, MSE at each time, including both bias and variance of estimated reliability metrics for each method are analyzed to investigate how much both bias and variance contribute the SSE. From the MSE analysis, MSE provides reliability estimation trend for each method. Parametric estimation method provides more accurate reliability estimation results than the other methods for most of sample sizes.

• The Korean Journal of Applied Statistics
• /
• v.24 no.2
• /
• pp.425-433
• /
• 2011

Reject inference in credit scoring is a statistical approach to adjust for nonrandom sample bias due to rejected applicants. Function estimation approaches are based on the assumption that rejected applicants are not necessary to be included in the estimation, when the missing data mechanism is missing at random. On the other hand, the density estimation approach by using mixture models indicates that reject inference should include rejected applicants in the model. When mixture models are chosen for reject inference, it is often assumed that data follow a normal distribution. If data include missing values, an application of the normal mixture model to fully observed cases may cause another sample bias due to missing values. We extend reject inference by a multivariate normal mixture model to handle incomplete characteristic variables. A simulation study shows that inclusion of incomplete characteristic variables outperforms the function estimation approaches.

• Journal of Astronomy and Space Sciences
• /
• v.27 no.2
• /
• pp.123-128
• /
• 2010

This paper analyzes that the global positioning system (GPS) receiver differential code bias (DCB) has effect on the estimation the ionosphere total electron content (TEC). The data from nine permanent GPS sites of the Korea Astronomy and Space Science Institute (KASI) were used for the estimation of the receiver DCB before (Trimble 4000 SSi) and after (Trimble NetRS) the receiver replacement, using the singular value decomposition method. The results showed that the estimated mean value of the receiver DCB varied from 0.11 ns (nanosecond) to 7.54 ns before the receiver replacement, but the receiver DCBs shoed large values than 20 ns except some stations after the replacement. The receiver DCB showed a relatively large difference by types of the receivers, and, as a result, it had a great effect on the estimation the ionosphere TEC using GPS.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.4
• /
• pp.1159-1165
• /
• 2005

Basu et al. (1998) proposed a new density-based estimator, called the minimum density power divergence estimator (MDPDE), which avoid the use of nonparametric density estimation and associated complication such as bandwidth selection. Woodward et al. (1995) examined the minimum Hellinger distance estimator (MHDE), proposed by Beran (1977), in the case of estimation of the mixture proportion in the mixture of two normals. In this article, we introduce the MDPDE for a mixture proportion, and show that both the MDPDE and the MHDE have the same asymptotic distribution at a model. Simulation study identifies some cases where the MHDE is consistently better than the MDPDE in terms of bias.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.7
• /
• pp.3229-3232
• /
• 2012

In this study, cancer incidence data were assessed to provide various rates of five year age groups for a given year, lying between two census years. The individual exponential growth rate method is most useful in both population-based and non-population cased cancer registries in India to estimate the population by five yearly age groups and also find the rates of crude rates, age standard rates and cumulative rates. This method has been shown to endure from bias and often results sacrificing the overall growth rate and correction factor must be needful in five year age group population to maintain it. A second method, the difference distribution method is also able to maintain the overall growth rate and overcome the bias in estimation of five yearly age group populations. From this point of view these methods serving a new technique for population estimation by five yearly age groups for inter census years.

• Communications for Statistical Applications and Methods
• /
• v.29 no.2
• /
• pp.263-275
• /
• 2022

Use of appropriate technique for non-response occurring in sample survey improves the accuracy of the estimation. Many studies have been conducted for handling non-ignorable non-response and commonly the response probability is estimated using the propensity score method. Recently, post-stratification method to obtain the response probability proposed by Chung and Shin (2017) reduces the effect of bias and gives a good performance in terms of the MSE. In this study, we propose a new response probability estimation method by combining the propensity score adjustment method using the logistic regression model with post-stratification method used in Chung and Shin (2017). The superiority of the proposed method is confirmed through simulation.

• The Korean Journal of Applied Statistics
• /
• v.33 no.4
• /
• pp.409-420
• /
• 2020

In business survey, modified cut-off sampling is commonly used to greatly increase the accuracy of the estimation while reducing the number of samples. However, non-response rate of take-all stratum has increased significantly and the sample substitution is not possible because the non-response in the take-all stratum affects the accuracy of the estimation. It is important to adjust the bias appropriately if non-response is affected by the variable of interest. In this study, a bias adjusted estimation is proposed as an appropriate method to deal with a non-response in the take-all stratum. In particular, the estimator proposed by Chung and Shin (2020) was applied to the bias adjustment for the take-all stratum; therefore, we suggest a new method to adjust properly for the take-all stratum. The superiority of the proposed estimator was examined through simulation studies and confirmed through actual data analysis.