• Communications for Statistical Applications and Methods
• /
• 제31권3호
• /
• pp.349-363
• /
• 2024

Compositional data refers to data where the sum of the values of the components is a constant, hence the sample space is defined as a simplex making it impossible to apply statistical methods developed in the usual Euclidean vector space. A natural approach to overcome this restriction is to consider an appropriate transformation which moves the sample space onto the Euclidean space, and log-ratio typed transformations, such as the additive log-ratio (ALR), the centered log-ratio (CLR) and the isometric log-ratio (ILR) transformations, have been mostly conducted. However, in scenarios with sparsity, where certain components take on exact zero values, these log-ratio type transformations may not be effective. In this work, we mainly suggest an alternative transformation, that is the square-root transformation which moves the original sample space onto the directional space. We compare the square-root transformation with the log-ratio typed transformation by the simulation study and the real data example. In the real data example, we applied both types of transformations to the USG% data obtained from NBA, and used a density based clustering method, DBSCAN (density-based spatial clustering of applications with noise), to show the result.

• 전기학회논문지
• /
• 제65권6호
• /
• pp.1019-1025
• /
• 2016

This paper proposes a parallel reduced-order square-root unscented Kalman filter for state estimation of a sensorless permanent-magnet synchronous motor. The appearance of an unscented Kalman filter is caused by the linearization process error between a real system and classical Kalman model. The unscented transformation can make a more accurate Kalman model. However, the complexity is its main drawback. This paper investigates the design and implementation of the proposed filter with Potter and Carlson square-root form. The proposed parallel reduced-order square-root unscented Kalman filter reduces memory and code size, and improves numerical computation. And the performance is not significantly different from the unscented Kalman filter. The experimentation is performed for the verification of the proposed filter.

• 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.

• 한국관개배수논문집
• /
• 제5권1호
• /
• pp.20-31
• /
• 1998

The Power transformation, the modified Power transformation, the logarithmic transformation, the square-root logarithmic transformation, the SMEMAX transformation, the Extreme value type III, the Weibull, the log Pearson type III, the lognormal distributi

• 한국농공학회지
• /
• 제32권E호
• /
• pp.20-32
• /
• 1990

Abstract This study was conducted to pursue the normalization of frequency distribution by making an approach to the coefficient of skewness to nearly zero through the Box-Cox transformation, to get probable flood flows can be calculated by means of the transformation equation which has been derivated by Box-Cox transformation in the annual maximum series of the applied watersheds. It has been concluded that Box-Cox transfromation is proved to be more efficient than logarithmic, square root and SMEMAX transformation which is based on the trigonometric solution of a right triangle whose three verteces repesent the smallest, median and largest observed values of a population in making the coefficient of skewness nearer to zero. Consequently it is shown that probable flood flows according to the return period based on Box-Cox transformation are closer to the observed data as compared to other methods including SMEMAX transformation and fitted probability distributions such as the three parameter lognormal and the type I extremal distribution for the applied watersheds.

• 대한물리치료과학회지
• /
• 제10권1호
• /
• pp.83-89
• /
• 2003

An electromyogram (EMG) using surface electrodes is one of the indirect tests most frequently used to ascertain muscle fatigue. An EMG can be used in two ways. The first technique determines the root mean square (RMS), which reflects the amplitude of the EMG signal. The second technique determines the median and mean power frequencies through EMG power spectrum analysis. The purpose of this article is for determine the correlation between work and percent root mean square(%RMS) and between work and MDF of EMG based on muscle contractions. It is used the %RMS, which reflects the amplitude of the EMG signal For MDF, it is used the frequency power spectrum analysis method, which involves the fast Fourier transformation (FFT) of the original Signals.

• 한국측량학회지
• /
• 제27권3호
• /
• pp.303-310
• /
• 2009

지구기준계 실현을 위해 최적망 구성 알고리즘에 근거하여 선택된 142개의 IGS 관측소를 선정하고 7년간 GPS 자료를 처리하였다. 본 연구에서 결정된 기준좌표와 속도를 이용하여 본 연구의 지구기준좌표계와 ITRF2005와의 변환매개변수를 추정하였으며, 각각의 관측소에서의 좌표와 속도를 비교하였다. 산출한 변환매개변수는 최대 4.3mm 이내에 분포하였고, 좌표와 속도 차의 RMS(root mean square)는 수평성분이 6.7mm와 1.3mm/yr, 수직성분이 13.3mm와 2.4mm/yr로서 ITRF2005와 잘 일치하고 있다. 이러한 연구는 측지기준계 결정을 위한 자체기술개발에 의의를 가지며 판구조운동 연구와 같은 전지구 관측을 위한 연구에 활용될 수 있다. 향후 다양한 자료처리 전략 개발과 더불어 관측소의 개수와 관측기간을 확대함으로써 보다 향상된 지구기준좌표계가 결정될 것이다.

• Journal of the Korean Data and Information Science Society
• /
• 제27권6호
• /
• pp.1465-1475
• /
• 2016

안전한 급식을 위해서는 식중독사고와 연관성이 있는 미생물의 위해요소들이 실시간적으로 모니터링 되고 통제되어야 한다. 선진국에서는 실시간 위생 모니터링 도구로 ATP (adenosine triphosphate) Luminometer를 활용한 사례가 여러 건 보고되었다. ATP analyser는 ATP bioluminescence (생물발광성)의 원리를 이용하여 RLU (relatively light unit) 값으로 위생수준을 간접적으로 측정할 수 있게 해준다고 알려져 있다. 이에 국내 급식산업에서도 이를 활용할 수 있도록 조리도구들을 대상으로 일반세균수 (aerobic plate count; APC)와 RLU 간에 상관성이 존재하는지를 확인 할 필요성이 제기 되었다. 본 연구는 급식소의 조리도구 표면에 사전 처리없이 ATP (RLU)와 APC (CFU)를 측정하여 상관관계 존재 여부 파악 및 이들 관계의 최적인 모델을 찾아보고자 하였다. 이들에 대한 분석 결과를 조리도구별로 요약하면 다음과 같다. 도마, 칼, 국그릇 (스텐), 식판 (카보) 자료는 1차 변환의 로그변환이, 컵 자료는 2차 변환의 제곱근-역변환 (혹은 역-제곱근변환)이, 국그릇 (카보) 자료는 2차 변환의 제곱근-제곱근변환이 표준화 회귀계수 및 결정계수 $R^2$가 가장 좋게 나타났으나 식판 (스텐) 자료는 원자료, 1차 변환 및 2차 변환 모든 경우에서 정규성을 만족하지 못하여 이번 자료에서는 최적인 경우를 찾을 수가 없었다.

• 한국측량학회지
• /
• 제17권3호
• /
• pp.257-264
• /
• 1999

본 연구는 근거리 수치사진의 카메라 렌즈방사왜곡의 보정에 있어서, 기존 상용 소프트웨어의 최소제곱법에 의한 좌표변환방법인 기하보정툴을 사용하지 않고, 주변환식으로는 카메라 검정자료에서 제공되는 고차다항식을, 역변환에 있어서는 Newton-Raphson의 비선형 방정식의근의 해법을 직접 사용하여 렌즈방사왜곡을 보정함으로써 기하학적 정확도가 최소제곱법에 의한 좌표변환방법에 비하여 약 0.04∼0.08화소 향상된 수치 사진화상을 얻을 수 있었다. 또한 기존 소프트웨어에 있어서는 고차다항식의 최소제곱법 적용을 위해서 동일한 경중률의 많은 기준점의 선정작업을 필요로 하나, 본 연구에서 제시한 기법은 이 작업이 필요 없으므로 작업의 효율성을 높일 수 있었다.

• 열처리공학회지
• /
• 제4권3호
• /
• pp.39-46
• /
• 1991

The effects of austenitizing temperatures and times and cooling rate on the characteristics of continuous air cooling have been investigated for 0.3%C-Mn steels microalloyed with vanadium. Transformation start temperatures have been found to be measured from temperature-time curve directly obtained with continuous air cooling and to decrease with increasing austenitizing temperature, cooling rate and Mn contents. The coarsening behavior of austenite grain size has been measured to abnormally grow at $1050^{\circ}C$ and rapidly grow at $1200^{\circ}C$. It has been found that the volume fraction of pearlite was linealy proportional to the reciprocal square root of austenite grain size. The hardness has been measured to increase with increasing cooling rate up to $250^{\circ}C/min.$ and to remain relatively unchanged in the range of $250{\sim}400^{\circ}C/min.$ showing that hardness valves for steel with a higher Mn content increase more than those for steel with a lower Mn content. The impact property has been found to decrease with increasing of austenite grain size but does not linealy change with the reciprocal square root of austenite grain size.