• 응용통계연구
• /
• 제28권3호
• /
• pp.511-527
• /
• 2015

ARMA(p, q) 모형 분석에서 분산 안정화 또는 정규화를 위해 멱변환(power transformation)이 사용된다. 변환된 자료를 이용하여 분석이 이루어지며 원 자료의 예측을 위해 재변환이 사용된다. 이때 흔히 변환된 자료 분석에서 얻어진 예측값의 역함수 값이 원자료 예측값으로 사용되지만 이는 편향이 있는 것으로 알려져 있다. 이를 해결하기 위해 로그 변환의 경우 Granger과 Newbold (1976)는 로그-정규분포의 기댓값을 이용할 것을 제안하였다. 본 연구에서는 모의실험을 통하여 제곱근 변환과 로그 변환 후 재변환을 사용할 때 예측값으로 기댓값의 역함수를 이용하는 방법과 역함수의 기댓값을 사용하였을 때의 추정의 결과를 모의실험을 통하여 비교하였다.

• 한국컴퓨터정보학회논문지
• /
• 제21권3호
• /
• pp.39-45
• /
• 2016

In this paper, we propose a log management service model for efficient developments of android applications. The proposed model consists of two major parts which are the log collector and the log manager service. The log collector can capture the log information of a target application without modifications, because the collector is composed by aspect-oriented programming. The collected logs are transformed to chunk of data, and the chunk of data is sent to the log management service. The log management service is an android service component and an independent application in another process. So, the log management service can reduce the workload of logging in the target application. Through a case study, we show that the proposed log management service model can reduce the log processing time compared to other models without modifications of a target application.

• 약학회지
• /
• 제44권4호
• /
• pp.308-314
• /
• 2000

Logarithmic transformation of pharmacokinetic parameters is routinely used in bioequivalence studies based on pharmacokinetic and statistical grounds by the United States Food and Drug Administration (FDA), European Committee for Proprietary Medicinal Products (CPMP), and Japanese National Institute of Health and Science (NIHS). Although it has not yet been recommended by the Korea Food and Drug Administration (KFDA), its use is becoming increasingly necessary in order to harmonize with international standards. In the present study, statistical procedures for the analysis of a bioequivalence based on the log transformation and a related SAS procedure were demonstrated in order to aid the understanding and application. The AUC parameters used in this demonstration were taken from the previous bioequivalence study for two aceclofenac tablets, which were performed in a single-dose crossover design. Analysis of variance (ANOVA), statistical power to detect 20% difference between the tablets, minimum detectable difference and confidence intervals were all assessed following log-transformation of the data. Bioequivalence of two aceclofenac tablets was then estimated based on the guideline of FDA. Considering the international effort for harmaonization of guidelines for bioequivalence tests, this approach may require a further evaluation for a future adaptation in the Korea Guidelines of Bioequivalence Tests (KGBT).

• 한국대기환경학회지
• /
• 제22권5호
• /
• pp.614-626
• /
• 2006

Using the data of three environmental monitoring sites in Pohang area(KME112, KME113, and KME114), statistical forecasting models of the daily maximum and mean values of PM10 have been developed. Since the distributions of the daily maximum and mean PM10 values are skewed, which are similar to the Weibull distribution, these values were log-transformed to increase prediction accuracy by approximating the normal distribution. Three statistical forecasting models, which are regression, neural networks(NN) and support vector regression(SVR), were built using the log-transformed response variables, i.e., log(max(PM10)) or log(mean (PM10)). Also, the forecasting models were validated by the measure of RMSE, CORR, and IOA for the model comparison and accuracy. The improvement rate of IOA before and after the log-transformation in the daily maximum PM10 prediction was 12.7% for the regression and 22.5% for NN. In particular, 42.7% was improved for SVR method. In the case of the daily mean PM10 prediction, IOA value was improved by 5.1% for regression, 6.5% for NN, and 6.3% for SVR method. As a conclusion, SVR method was found to be performed better than the other methods in the point of the model accuracy and fitness views.

• 한국수자원학회논문집
• /
• 제41권8호
• /
• pp.807-823
• /
• 2008

수위-유량 관계 곡선을 나타내는 곡선식에 포함되어 있는 매개변수의 추정을 위해 많이 사용되는 로그선형 회귀분석은 잔차의 비등분산성(heteroscedasticity)을 고려하지 못하므로 본 연구에서는 의사우도추정법(pseudolikelihood estimation, P-LE)에 의해 분산함수를 추정하고 이와 함께 회귀계수를 추정할 수 있는 방법을 제시하였다. 이 과정에서 제시된 회귀잔차를 최소화하기 위하여 SA(simulated annealing)이라는 전역 최적화 알고리즘을 적용하였다. 또한 수위-유량 관계 곡선은 단면 등의 영향으로 인해 구간에 따라 각각 다르게 구축되어져야 하므로 이를 보다 객관적으로 판단하고 분리 위치를 추정하기 위하여 Heaviside 함수를 의사우도함수에 포함시켜 결과를 추정하도록 하였으며, 2개의 구간을 가지는 유량자료를 이용하여 제시된 방법의 합리성을 통계적으로 실험하였다. 이와 같이 통계적 실험을 통해 제시된 방법들이 기존 방법과 비교하여 가질 수 있는 장점을 파악하였으며, 제시된 방법들을 금강유역 5개 지점에서 대해 수행하여 효율성을 검증하였다.

• Communications for Statistical Applications and Methods
• /
• 제31권4호
• /
• pp.409-425
• /
• 2024

Logistic regression models are commonly used to explain binary health outcome variable using independent variables such as patient characteristics in medical science and public health research. Although there is no distributional assumption required for independent variables in logistic regression, variables with severely right-skewed distribution such as lab values are often log-transformed to achieve symmetry or approximate normality. However, lab values often have zeros due to limit of detection which makes it impossible to apply log-transformation. Therefore, preprocessing to handle zeros in the observation before log-transformation is necessary. In this study, five methods that remove zeros (shift by 1, shift by half of the smallest nonzero, shift by square root of the smallest nonzero, replace zeros with half of the smallest nonzero, replace zeros with the square root of the smallest nonzero) are investigated in logistic regression setting. To evaluate performances of these methods, we performed a simulation study based on randomly generated data from log-normal distribution and logistic regression model. Shift by 1 method has the worst performance, and overall shift by half of the smallest nonzero method, replace zeros with half of the smallest nonzero method, and replace zeros with the square root of the smallest nonzero method showed comparable and stable performances.

• Journal of Information Processing Systems
• /
• 제7권4호
• /
• pp.613-626
• /
• 2011

In this paper, we focus on the pinwheel task model with a variable voltage processor with d discrete voltage/speed levels. We propose an intra-task DVS algorithm, which constructs a minimum energy schedule for k tasks in O(d+k log k) time We also give an inter-task DVS algorithm with O(d+n log n) time, where n denotes the number of jobs. Previous approaches solve this problem by generating a canonical schedule beforehand and adjusting the tasks' speed in O(dn log n) or O($n^3$) time. However, the length of a canonical schedule depends on the hyper period of those task periods and is of exponential length in general. In our approach, the tasks with arbitrary periods are first transformed into harmonic periods and then profile their key features. Afterward, an optimal discrete voltage schedule can be computed directly from those features.

• 한국농공학회지
• /
• 제30권4호
• /
• pp.134-153
• /
• 1988

This paper was carried out to investigate the existence of a unique stress- strain behavior by obtaining some factors influencing the time dependent stress- strain behavior of clay. The results obtained from this study were summarized as follows ; 1. The relationship between stress ratro and strain in normally consolidated clay was in- dependent on pre-shear consolidation pressure. Therefore, shear strain could be expressed as a function with stress ratio. 2. The constitutive equation of shear strain on Modified Carn Clay Model coincided better with the observed value than Cam Clay Model. 3. The relationships between deviator stress and shear strain, between pore water pressure and shear strain were unified by the mean equivalent pressure. 4. The shear strain contour in norrnally consolidated clay was increased linearly through origin, but that in overconsolidated clay was not in accordance with the result of the former. 5. Because the effective stress path of normally consolidated clay was unified by the mean equivalent pressure, state boundary surface in (e,p,q) space was transformed into two dimensional surface. But it was considered to be suitable that the unified stress- strain in overconsolidated clay be expressed by a function with overconsolidation ratio. 6. The deviator for constant strain was increased linearly with increment of strain rate ($\varepsilon$) on semi-log scale, but pore water pressure was decreased. 7. The behavior of stress relaxation was transformed from linear to curvilinear with inc - rement of strain rate before stress relaxation test, and pore water pressure was increased in total range. 8. The strain of creep was increased linearly with increment of time on semi-log scale. The greater the strain rate before creep test became, the greater the increment of strain of creep became. And the pore water pressure during creep test was increased generally with increment of time on semi-log scale.

• ETRI Journal
• /
• 제46권1호
• /
• pp.22-34
• /
• 2024

Exposure to varied noisy environments impairs the recognition performance of artificial intelligence-based speech recognition technologies. Degraded-performance services can be utilized as limited systems that assure good performance in certain environments, but impair the general quality of speech recognition services. This study introduces an audiovisual speech recognition (AVSR) model robust to various noise settings, mimicking human dialogue recognition elements. The model converts word embeddings and log-Mel spectrograms into feature vectors for audio recognition. A dense spatial-temporal convolutional neural network model extracts features from log-Mel spectrograms, transformed for visual-based recognition. This approach exhibits improved aural and visual recognition capabilities. We assess the signal-to-noise ratio in nine synthesized noise environments, with the proposed model exhibiting lower average error rates. The error rate for the AVSR model using a three-feature multi-fusion method is 1.711%, compared to the general 3.939% rate. This model is applicable in noise-affected environments owing to its enhanced stability and recognition rate.

• 대한원격탐사학회지
• /
• 제21권3호
• /
• pp.173-188
• /
• 2005

The present study aims to derive and compare narrow and broad bandwidths of ocean color sensor’s algorithms for the study of monitoring highly dynamic coastal oceanic environmental parameters using high-resolution imagery acquired from Multi-spectral Camera (MSC) on KOMPSAT-2. These algorithms are derived based on a large data set of remote sensing reflectances ($R_{rs}$) generated by using numerical model that relates $b_b/(a + b_b)$ to $R_{rs}$ as functions of inherent optical properties, such as absorption and backscattering coefficients of six water components including water, phytoplankton (chl), dissolved organic matter (DOM), suspended sediment (SS) concentration, heterotropic organism (he) and an unknown component, possibly represented by bubbles or other particulates unrelated to the first five components. The modeled $R_{rs}$ spectra appear to be consistent with in-situ spectra collected from Korean waters. As Kompsat-2 MSC has similar spectral characteristics with Landsat-5 Thematic Mapper (TM), the model generated $R_{rs}$ values at 2 ㎚ interval are converted to the equivalent remote sensing reflectances at MSC and TM bands. The empirical relationships between the spectral ratios of modeled $R_{rs}$ and chlorophyll concentrations are established in order to derive algorithms for both TM and MSC. Similarly, algorithms are obtained by relating a single band reflectance (band 2) to the suspended sediment concentrations. These algorithms derived by taking into account the narrow and broad spectral bandwidths are compared and assessed. Findings suggest that there was less difference between the broad and narrow band relationships, and the determination coefficient $(r^2)$ for log-transformed data [ N = 500] was interestingly found to be $(r^2)$ = 0.90 for both TM and MSC. Similarly, the determination coefficient for log-transformed data [ N = 500] was 0.93 and 0.92 for TM and MSC respectively. The algorithms presented here are expected to make significant contribution to the enhanced understanding of coastal oceanic environmental parameters using Multi-spectral Camera.