• Communications for Statistical Applications and Methods
• /
• v.28 no.5
• /
• pp.563-582
• /
• 2021

This paper deals with developing a general class of accelerated sequential procedures and obtaining the associated second-order approximations for the expected sample size and 'regret' (difference between the risks of the proposed accelerated sequential procedure and the optimum fixed sample size procedure) function. We establish that the estimation problems based on various lifetime distributions can be tackled with the help of the proposed class of accelerated sequential procedures. Extensive simulation analysis is presented in support of the accuracy of our proposed methodology using the Pareto distribution and a real data set on carbon fibers is also analyzed to demonstrate the practical utility. We also provide the brief details of some other inferential problems which can be seen as the applications of the proposed class of accelerated sequential procedures.

• The Korean Journal of Applied Statistics
• /
• v.14 no.1
• /
• pp.13-24
• /
• 2001

In this paper, we investigated mean estimation methods in two-phase sampling. Under the fixed expected cost we reviewed the optimal sample sizes, minimum variances and approximate unbiased variance estimators for usual ratio estimator, stratified sample mean with proportional allocation and Rao's allocation of the second phase sample. Also we proposed combined ratio estimator, which uses both ratio estimation and stratification and derived optimal sample size, minimum variance and unbiased variance estimator. Through a limited simulation study, we compared estimators by design effects and came to know that ratio estimator is more efficient than stratified sample mean in some cases and inefficient in the other cases, but combined ratio estimator is more efficient than others in most cases.

• Journal of the Korean Wood Science and Technology
• /
• v.49 no.5
• /
• pp.491-503
• /
• 2021

Wood identification is regularly performed by observing the wood anatomy, such as colour, texture, fibre direction, and other characteristics. The manual process, however, could be time consuming, especially when identification work is required at high quantity. Considering this condition, a convolutional neural networks (CNN)-based program is applied to improve the image classification results. The research focuses on the algorithm accuracy and efficiency in dealing with the dataset limitations. For this, it is proposed to do the sample selection process or only take a small portion of the existing image. Still, it can be expected to represent the overall picture to maintain and improve the generalisation capabilities of the CNN method in the classification stages. The experiments yielded an incredible F1 score average up to 93.4% for medium sample area sizes (200 × 200 pixels) on each CNN architecture (VGG16, ResNet50, MobileNet, DenseNet121, and Xception based). Whereas DenseNet121-based architecture was found to be the best architecture in maintaining the generalisation of its model for each sample area size (100, 200, and 300 pixels). The experimental results showed that the proposed algorithm can be an accurate and reliable solution.

• Journal of Korean Society for Quality Management
• /
• v.20 no.1
• /
• pp.48-58
• /
• 1992

This paper deals with a test procedures on reliablity using hybrid censoring when failure time follows two parameter Weibull distribution. In each case of single and two stage test with hybrid censoring, we construct a operating characteristic curve, and then obtain the censoring number and sample size which the producer's risk and the consumer's risk are both satisfied. This study suggests to determine the expected waiting time at a decision.

• Journal of the Korean Statistical Society
• /
• v.29 no.4
• /
• pp.435-441
• /
• 2000

Ko(1998, 1999) proposed certain flexible two-stage plans that could be served as one-step interim analysis in on-going clinical trials. The proposed Plans are optimal simultaneously in both a Bayes and a Neyman-Pearson sense. The Neyman-Pearson interpretation is that average expected sample size is being minimized, subject just to the two overall error rates $\alpha$ and $\beta$, respectively of first and second kind. The Bayes interpretation is that Bayes risk, involving both sampling cost and wrong decision losses, is being minimized. An example of this correspondence are given by using a binomial setting.

• Journal of the Korean Statistical Society
• /
• v.16 no.1
• /
• pp.37-44
• /
• 1987

A two-stage selection procedure is considered in the case of exponential populations with common known scale parameter. The proposed procedure is designed following the lines of Tamhane and Bechhofer(1977). The design constants to implement the procedure are provided. Monte Carlo results show that the proposed procedure performs better than the single procedure by Raghvachari and Starr (1970) in terms of the expected total sample size.

• Journal of Distribution Science
• /
• v.12 no.5
• /
• pp.61-69
• /
• 2014

Purpose - Many studies report that returns on hedge portfolios that eliminate particular risk types are abnormal from traditional asset pricing models' perspectives. This study examines the pervasiveness of anomalous returns conditioned on business cycle and group size. Research design, data, and methodology - Using KOSPI and KOSDAQ market data from July 1991 to December 2013, we categorize stocks into appropriately sized groups, and dichotomize our sample periods into expansion and recession periods then, we construct hedge portfolios by sorting stocks by anomaly variables and calculate their returns. Results - Four anomalies, including earnings yield, net stock issue, total asset growth, and liquidity appear pervasive across all groups for the entire sample period. However, only the hedge returns of net stock issues are significant across all group sizes during both expansion and recession. Conclusions - A net stock issue can be an appropriate proxy for expected growth of book equity for all group sizes in recessions. This finding could provide insights to investment industry participants and to researchers interested in the relationship between expected growth of book equity and business cycle risk.

• Journal of Applied Reliability
• /
• v.14 no.4
• /
• pp.225-229
• /
• 2014

A one-shot device is defined as a product, system, weapon, or equipment that can be used only once. After use, the device is destroyed or must undergo extensive rebuild. Determining the reliability of a one-shot device poses a unique challenge to the manufacturers and users due to the destructive nature and costs of the testing. This paper presents a reliability growth prediction for a one-shot system. It is assumed that 1) test duration is discrete(i.e. trials or rounds); 2) trials are statistically independent; 3) the number of failures for a given system configuration is distributed according to a binomial distribution; and 4) the cumulative expected number of failures through any sequence of configurations is given by AMSAA model. When the system development is represented by three configurations and the number of trials and failures during configurations are given, the AMSAA model parameters and reliability at configuration 3 are estimated by using a reliability growth analysis software. Further, if the reliability growth predictions do not meet the target reliability, the sample size of an additional test is determined for achieving the target reliability.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.2
• /
• pp.68.2-68.2
• /
• 2014

High Frequency Peakers (HFPs) are radio-loud Active Galactic Nuclei (AGN), which are regarded as being in the earliest evolutionary phase (102-103 years) of radio galaxies. They are expected to be small in size (< ~1 kpc) compared to their host galaxies (~a few 10s kpc), and have convex spectra, which are peaking at high radio frequency (> 5 GHz). Their size and spectral shape are the most obvious supporting evidence of extremely young ages. HFPs are therefore ideal targets to probe the earliest stage of radio sources. To date however, the young radio source classification has been relying mainly on the spectral shape which usually does not cover high enough frequencies where the true peak flux is located. Hence HFPs are often confused with blazars which may show a similar spectral shape and apparent compactness but are a somewhat evolved form of AGNs. Therefore, we have been challenging to identify HFPs among the sample of 19 candidates using the Korean VLBI Network (KVN) which enables us to extend the radio spectrum baseline up to 22 and 43 GHz. These are higher than the frequencies used in most previous studies of HFPs, allowing us to select genuine HFPs. By long-term monitoring of 18 epochs, we have also inspected the variability of the sample to select out blazars which are highly variable yet with a similar radio spectrum. In this work, we present the light curves and spectral properties of the HFP candidates. We discuss the results of our re-identification of HFPs.

• Communications for Statistical Applications and Methods
• /
• v.12 no.1
• /
• pp.29-38
• /
• 2005

In sample design, determining the weights of estimates becomes usually great influence on the result. In this article, raking methods are applied to different domain and depending on the range of the domain and sample size, the results of estimates are explained and compared. For the comparison, we use the MSE, MAE, MSPE and MAPE with Actual State of Minor Enterprisers Human Resources Survey data in 2001. The simulation result shows that more elaborate method is superior to the widely used method as expected but the difference is not quite significant.