• 대한원격탐사학회지
• /
• 제18권3호
• /
• pp.155-162
• /
• 2002

Neural networks have been focused on as a robust classifier for the remotely sensed imagery due to its statistical independency and teaming ability. Also the artificial neural networks have been reported to be more tolerant to noise and missing data. However, unlike the conventional statistical classifiers which use the statistical parameters for the classification, a neural network classifier uses individual training sample in teaming stage. The training performance of a neural network is know to be very sensitive to the discrepancy of the number of the training samples of each class. In this paper, the effect of the population discrepancy of training samples of each class was analyzed with three layered feed forward network. And a method for reducing the effect was proposed and experimented with Landsat TM image. The results showed that the effect of the training sample size discrepancy should be carefully considered for faster and more accurate training of the network. Also, it was found that the proposed method which makes teaming rate as a function of the number of training samples in each class resulted in faster and more accurate training of the network.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권5호
• /
• pp.179-188
• /
• 2021

The purpose of this study is to examine the level of aggressive tax planning (ATP) among companies listed in the Access, Certainty, Efficiency (ACE) Market of Bursa Malaysia. On top of that, this study also investigates the relationship between company characteristics, ethnicity, and ATP. This study uses a balanced pooled sample of 105 firm years-observations for the period from 2014 to 2018. These samples were selected to provide new insight into this market and to explore the attitude of small firms toward ATP in Malaysia. The data was retrieved from DataStream and the downloaded annual reports. The finding shows that profitability and financial distress have a significant relationship with ATP. Other variables including size, capital intensity, inventory intensity, leverage, and ethnicity, were not determinants of ATP. The result in this study may assist the reader in understanding the nature of companies in the ACE market, particularly on its behavior toward tax planning. A strict requirement is needed to be adopted in the sample selection process, thus limiting the sample size. Further, since the previous study focused on large companies, the discussion of this paper will provide new insight into the nature of tax planning within the small- and medium-sized companies in Malaysia.

• 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.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권2호
• /
• pp.143-150
• /
• 2021

This study investigates and analyzes the difference in Indonesian banks' earnings quality in the pre-audit and post-audit period. This study also investigates the difference in audit quality done by public accounting firms. This study employs time series data taken from the unaudited and audited financial statements of banks listed on the Indonesia Stock Exchange in 2012-2016. Sample selection is made by using a purposive sampling method. The population of this study is 43 banks, and after checking the data for validity and reliability, the final sample size was 26 banks. Audit quality is operationalized with the size of the auditor. Earnings quality is proxied by accruals calculated using the Beaver and Engel (1996) model. The data analysis method used in this study is the paired-sample t-test and chow test. This study shows that there is no difference in earnings quality in the pre-audit and post-audit period. This study also reveals no difference in audit quality between the big four and non-big four auditors. These findings mean that independent auditors do not play a useful role in increasing the reliability of accounting information presented by management to stakeholders. Besides, this study's results do not verify the agency theory regarding auditors' role to minimize opportunistic management behavior in preparing financial statements.

• 산경연구논집
• /
• 제13권11호
• /
• pp.11-19
• /
• 2022

Purpose: Food security is a dynamic concept, which has continuously integrated new dimensions and levels of analysis over the years; this reflects the wider recognition of its complexities in research and public policy issues. Food insecurity is a major challenge for Ethiopia and for many African countries. This paper aims to identify the roles of social enterprises to overcome food insecurity. Research design, data and methodology: The research has focused on one primary question. What are the role of social enterprises to overcome food insecurity? The survey questionnaire was developed and an interview was made using enumerators. A total sample size of 400 sample households was selected randomly using the method of sampling with probability proportionate to size. Descriptive and inferential statistical analysis methods were used for data analysis. Results: Results of the regression analysis show that all of the hypothesized independent variables are significantly and positively affecting post membership income of the sample households. Conclusions: all stakeholders as government, development actors and academic institutions should closely monitor and support social enterprises/cooperatives technically, financially and materially. By doing so, the full potential and contributions of social enterprises in reducing unemployment and solving the social and environmental problems can be effective.

• 국제지역연구
• /
• 제13권3호
• /
• pp.527-549
• /
• 2009

본 연구는 Chen and Zhao(2006)에 의해 실증된 M/B비율과 부채비율 간의 역 U자형 관계가 시장, 기업규모 및 기술수준별로 어떠한 차이를 보이는지를 검정하여 보았다. 이를 위해 2001년도 1월 1일부터 2008년도 12월 31일까지 한국 유가증권시장과 코스닥시장에 계속 상장된 510개 기업, 4,080개의 관측수를 확보하였다. 각 표본별 관측수는 유가증권시장표본이 2,448개이며, 코스닥시장표본이 1,632개이다. 기업규모별로는 대기업표본이 1,816개, 중소기업표본이 2,264개이며, 기술수준별로는 중 및 저기술업종표본이 1,704개, 첨단 및 고기술업종표본이 2,376개이다. 주요 실증분석결과는 다음과 같다. 첫째, 평균 M/B비율은 유가증권시장표본(대기업표본, 중 및 저기술업종표본)에 비해 코스닥시장표본(중소기업표본, 첨단 및 고기술업종표본)이 높았으며, 유상증자비율의 평균 역시 전자에 비해 후자가 높았다. 둘째, 코스닥시장표본(중소기업표본, 첨단 및 고기술업종표본)의 경우 M/B비율과 부채비율 간에 역 U자형 관계를 확인할 수 있었으나, 유가증권시장표본(대기업표본, 중 및 저기술업종표본)의 경우 이러한 관계를 찾아볼 수 없었다. 후자에 있어 역 U자형 관계가 나타나지 않는 것은 이들 기업의 낮은 M/B비율에 기인한 것으로 생각된다.

• 한국산업융합학회 논문집
• /
• 제26권1호
• /
• pp.183-191
• /
• 2023

This study was conducted experimentally to investigate the surface temperature of the heat sink, the air temperature in the flow channel and the sample temperature by changing the channel number of channel type heat sink and the air flow rate when heating and cooling the bio sample. The target temperature of the sample was 15℃ or less as the minimum value and 82℃ or more as the maximum value. In this study, the channel number of the heat sink(N = 1, 2, 4, 5, 10) and the air flow rate(Q=25, 42, 54m³/min) were varied. The bio sample was replaced with water, and the volume of water is 4mL. The size of the heat sink is 80x73x150mm and the material is aluminum. When cooling the sample, the surface temperature, the air temperature and the sample temperature were highly dependent on the number of channels and the flow rate. However, when the sample is heated, the surface temperature, air temperature and sample temperature do not depend on the number of channels and the flow rate. It was found that the conditions for satisfying the minimum temperature of 15℃ or less when cooling the sample were the number of channels N≥5 and the flow rate Q≥42m³/min. When heating the sample, the conditions to satisfy the maximum temperature of 82℃ or more are the number of channels N≤5 and the air flow rate Q≤42m³/min.

• Communications for Statistical Applications and Methods
• /
• 제7권2호
• /
• pp.563-573
• /
• 2000

In this thesis, we propose the test statistic for ordered alternatives on c-sample ranked set samples(RSS). The proposed test statistic JRSS is Jonckheere type statistic using the median of the i-th samples in each cycle. We obtained the asymptotic property of the proposed test statistic and the asymptotic relative efficiencies of the proposed test statistic with respect to J SRS which Jonckheere type statistic on simple random samples(SRS). From the simulation works, J RSS is superior to J SRS. We compared the empirical powers of J RSS with respect to U RSS on ranked set sample and U SRS on simple random sample using all samples, which are proposed by Kim, Kim and Lee(1999). The powers of J RSS are nearly the same values when entire sample size is large. J RSS is superior to U RSS. J RSS is simpler than U RSSon calculating process.

• 한국정밀공학회지
• /
• 제16권2호통권95호
• /
• pp.68-73
• /
• 1999

Generally, most of machining processes are done according to the dimention of the draft made by CAD. However, there are many cases that a sample is given without the draft because of the simplicity of the shape in the machining of 2D objects. To cut the same shape as the given sample, this paper proposes the method to extract the geometric information about a large sample using the robot vision and to draw the demensional draft for the machining. Because the resolution of one frame in the vision system is too low, it is necessary to set up a camera according to the desired resolution and to capture the image moving along the contour. And the overall outline can be compounded of the sequentially captured images. In the experiment, we compared the product after the cutting with the original sample and found that the size of two objects was coincided within the allowed error bound.

• Journal of the Korean Data and Information Science Society
• /
• 제8권1호
• /
• pp.71-77
• /
• 1997

S independent sample 0,1,2, $\cdots$, s-1 (or stages 0,1,2, $\cdots$, s-1) are available from the Raleigh population. Procedure for predicting any order statistic in the $(s+1)^{th}$ sample is developed by obtaining the predictive distribution at stage s. Bounds for the sample size at stage S, in order to have the variance at stage S less than that at stage (s-1), are obtained.