• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.386-389
• /
• 2008

It is important that the training stage of a supervised classification is designed to provide the spectral information. On the design of the training stage of a classification typically calls for the use of a large sample of randomly selected pure pixels in order to characterize the classes. Such guidance is generally made without regard to the specific nature of the application in-hand, including the classifier to be used. An approach to the training of a support vector machine (SVM) classifier that is the opposite of that generally promoted for training set design is suggested. This approach uses a small sample of mixed spectral responses drawn from purposefully selected locations (geographical boundaries) in training. A sample of such data should, however, be easier and cheaper to acquire than that suggested by traditional approaches. In this research, we evaluated them against traditional approaches with high-resolution satellite data. The results proved that it can be used small mixed pixels to derive a classification with similar accuracy using a large number of pure pixels. The approach can also reduce substantial costs in training data acquisition because the sampling locations used are commonly easy to observe.

• The Korean Journal of Applied Statistics
• /
• v.20 no.1
• /
• pp.103-115
• /
• 2007

In this paper we studied the small sample asymptotic inference for the autoregressive coefficient in AR(1) model. Based on saddlepoint approximations to the distribution of quadratic forms, we suggest a new approximation to the distribution of the estimators of the noncircular autoregressive coefficients. Simulation results show that the suggested methods are very accurate even in the small sample sizes and extreme tail area.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2000.06a
• /
• pp.57-73
• /
• 2000

Small area estimation is becoming important in survey sampling due to a growing demand for reliable small area statistics. In estimating means, totals, and other parameters for small areas of a finite population, samplie sizes for small areas are typically small because the overall sample size is usually determined to provide specific accuracy at a much higher level of aggregation than that of small area. The usual direct estimators that use the only information which is gotten from the sample in a given small area provide unreliable estimates. However, indirect estimators utilize the information from the areas related with a given small area, that is borrow strength from other related areas, and so give more accurate estimates than direct estimators. In this paper we investigate small area estimation methods such as synthetic, composite and empirical best linear unbiased prediction estimator, and apply them to real domestic data which is from the Survey of Hotels and Restaurants in In-Chon as of 1996 and then evaluate the performance of these methods by measuring average squared errors. This evaluation shows that indirect estimators, which are small area estimation methods, are more efficient direct estimator.

• The Journal of Korean Medicine
• /
• v.19 no.1
• /
• pp.179-204
• /
• 1998

The purpose of this research was to investigate the efficacy of Goiwhasan extract powder on the gastric injuries, antiulcer, gastrointestinal tract and blood hemostasis. Animals were used through this studies mice and rats. All animals were divided into 3 groups, contol group(no treatment), sample Ⅰ group(375mg/kg administration), sample Ⅱ group(750mg/kg administration). The gastric injuries and ulcer have been made by using pyloric ligation, indomethacin, HCI-ethanol, acetic acid and then The histological observation was followed. In the gastrointestinal tract, gastric juice secretion, gastric acidity, pepsin output, blood gastrin and secretin level, transport potentials in the small and large intestine were checked. And studies on blood hemostasis were performed on normal hemostatic activities and plasma prothrombin time, plasma recalcification time, plasma fibrinogen levels in the hypoprothrombinemic mice induced by warfarin. The results were as follows: 1. The antigastric ulcer effects on the pyloric ligation, indomethacin, HCl-ethanol, acetic acid induced gastric injuries were shown in Sample Ⅱ group(p<0.05). 2. Through the morphologic examination on the acetic acid induced ulcer, Sample Ⅰ group showed mild regeneration of epithelium and slight decrease of periulcer edema then that of Control group, while Sample Ⅱ group showed more retraction of round ulcer site, remarkable loss of swelling and edema then that of Control group, and revealing the regenerated epithelium in the surrounding ulcer site. Thus it was noted that both Sample groups have antigastriculcer effects on the experimentally induced gastric ulcer. 3. The inhibitory effects on gastric juice were noted in both Sample Ⅰ group(p<0.05) and Sample Ⅱ group(p<0.01). However, only Sample Ⅱ group showed the inhibitory effects on total acidity and pepsin output(p<0.05). 4. The significant inhibition of blood gastrin level showed at 30 min.(P<0.05) and 90 min.(P<0.05) after starting medication in only Sample Ⅱ group, but significance of blood secretin level in both groups was not recognized. 5. Any significant changes in barium sulfate transport in the small intestine of mice was not recognized in both groups, but the significantly inhibitory effect in large intestine was recognized in both Sample Ⅰ group(p<0.05) and Sample Ⅱ group(p<0.001). 6. In hemostatic effect on both normal mice and hypoprothrombinemic mice induced by warfarin, the significantly shortening effect on coagulation time was seen in only Sample Ⅱ group(p<0.01). 7. On plasma prothrombin time in hypoprothrombinemic rat induced by warfarin, Sample Ⅱ group have shortened the prothrombin time significantly(p<0.001). 8. On plasma reclcification time in hypoprothrombinemic rat induced by warfarin, the recalcification time have been shortened significantly in both Sample Ⅰ group(p<0.05) and Sample Ⅱ group(p<0.01). 9. On plasma fibrinogen levels in hypoprothrombinemic rat induced by warfarin, the fibrinogen contents in Sample Ⅱ have been decreased significantly(p<0.01). Overall the above results suggest that Goiwhasan has an therapeutic efficacy on antigastric ulcer and blood hemostasis. Further studies would be needed on the interaction of its herbal medicine and its mechanism in the future.

• Journal of Veterinary Clinics
• /
• v.32 no.1
• /
• pp.73-77
• /
• 2015

There has been increasing attention on sample size requirements in peer reviewed medical literatures. Accordingly, a statistically-valid sample size determination has been described for a variety of medical situations including diagnostic test accuracy studies. If the sample is too small, the estimate is too inaccurate to be useful. On the other hand, a very large sample size would yield the estimate with more accurate than required but may be costly and inefficient. Choosing the optimal sample size depends on statistical considerations, such as the desired precision, statistical power, confidence level and prevalence of disease, and non-statistical considerations, such as resources, cost and sample availability. In a previous paper (J Vet Clin 2012; 29: 68-77) we briefly described the statistical theory behind sample size calculations and provided practical methods of calculating sample size in different situations for different research purposes. This review describes how to calculate sample sizes when assessing diagnostic test performance such as sensitivity and specificity alone. Also included in this paper are tables and formulae to help researchers for designing diagnostic test studies and calculating sample size in studies evaluating test performance. For complex studies clinicians are encouraged to consult a statistician to help in the design and analysis for an accurate determination of the sample size.

• Journal of Veterinary Clinics
• /
• v.31 no.4
• /
• pp.288-292
• /
• 2014

A critical assumption of the standard sample size calculation is that the response (outcome) for an individual patient is completely independent to that for any other patient. However, this assumption no longer holds when there is a lack of statistical independence across subjects seen in cluster randomized designs. In this setting, patients within a cluster are more likely to respond in a similar manner; patient outcomes may correlate strongly within clusters. Thus, direct use of standard sample size formulae for cluster design, ignoring the clustering effect, may result in sample size that are too small, resulting in a study that is under-powered for detecting the desired level of difference between groups. This paper revisit worked examples for sample size calculation provided in a previous paper using nomogram to easy to access. Then we present the concept of cluster design illustrated with worked examples, and introduce design effect that is a factor to inflate the standard sample size estimates.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.8
• /
• pp.325-334
• /
• 2021

The study investigates the effectiveness of the macroprudential policy on credit growth in Vietnam. The authors use the logic of the transmission mechanism of macroprudential policy on credit growth. Research variables include economic growth, inflation, interest rate, and quarterly bank-level data from 28 commercial banks in Vietnam during 2011-2018. The results reveal that: (i) GDP growth had a positive impact on credit growth of small banks but had no impact on large banks, (ii) Domestic Systemically Important Banks (D-SIBs) and small banks respond differently to macroprudential measures of imposing different credit growth targets for different bank groups, (iii) Restrictions on foreign currency loans are found to be effective in curbing credit growth for the full sample and small banks, (iv) Inflation and economic cycle have significantly impacted credit growth at bank-level in Vietnam and (v) Interestingly, a significant positive relationship between interest rates and credit growth is found for the full sample and D-SIBs in Vietnam. The findings suggest that a stable macroeconomic environment should be good conditions for financial stability, and monetary authority should pay more attention to small banks' behaviors than D-SIBs behavior, toward such "administration" tools since small banks tend to prefer "breaking the rules" to make profits.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1125-1125
• /
• 2001

Micro-scale test methods are producing small-sample size where the conventional physical and chemical tests can not be used (high standard deviation, uncertain sampling conditions, low repeatability). Different small-scale test methods were developed recently for determination of physico-chemical, functional, rheological properties of wheat or wheat dough using miniaturized instruments with sophisticated sample preparation/handling and mechanics (RVA, 2 g mixograph, micro-Z-arm mixer, small-scale noodle maker, micro-baking method etc.). The small-scale methodologies can be used as basic research tools or as technology supported measurements and can be also essential in the early selection for quality traits in breeding programs. The milling as a sample preparation step is essential procedure providing good quality flour or semolina samples from small amount of grain (5-10 g) in a reproducible and reliable way. The aim of present study was to use NIR as quality control tool, and to evaluate the recently developed and manufactured micro-scale lab mill (FQC-2000) produced by Inter-Labor Co. Ltd., Hungary. The milling characteristics of the new instrument were compared to other laboratory mills and the effects of milling action on the chemical composition of fractions were analysed. The fractions were tested with both chemical and near infrared spectroscopic methods. The micro-scale milling resulted significantly different yields, particle size distributions and different fractions from compositional point of view. The near infrared spectra were sensitive enough to distinguish the fractions obtained by different milling procedures. Quantitative NIR calibration equations were developed and tested in order to measure the chemical composition of characteristic milling fractions. Special qualification procedure the PQS (Polar Qualification System) method was used for detecting the differences between fractions obtained by macro and micro-milling procedures. The results and the limitations of PQS method in this application will be discussed.

• Communications for Statistical Applications and Methods
• /
• v.17 no.3
• /
• pp.337-347
• /
• 2010

Most sample surveys are designed to estimate reliable statistics for the whole population and for some large subpopulations. However, the research for small area estimation have been increasing in recent years because users demand to reliable estimates for smaller subpopulations like small areas or specific domains. In Korea, the Economically Active Population Survey(EAPS) is the main household survey that produces monthly unemployment rates for nationwide and 16 large areas (7 metropolitans and 9 provinces) in Korea. For county level estimation, direct estimators are not reliable because of the small sample sizes. We consider small area estimation of the county level unemployment ratesfrom the sample observations in EAPS. To do this, we use an area level model to "borrow strength" from the auxiliary information, such as administrative data and census data. The proposed method is based on the assumption of normality of the model errors in the area level model. The proposed method is compared with the other alternatives in terms of the estimated mean squared errors.

• Communications for Statistical Applications and Methods
• /
• v.6 no.3
• /
• pp.659-666
• /
• 1999

In comparing two population means, the test statistic depends on the sample means and the variances, which are very sensitive to the extremely large or small values. This paper aims at examining the behavior of such observations using proper criterion which can measure the influence of them. We derive a computationally feasible statistic which can detect influential observations on the two-sample t-statistic.