The Experts below are selected from a list of 12096 Experts worldwide ranked by ideXlab platform
Tae Kyun Kim - One of the best experts on this subject based on the ideXlab platform.
-
T test as a Parametric Statistic
Korean Journal of Anesthesiology, 2015Co-Authors: Tae Kyun KimAbstract:In Statistic tests, the probability distribution of the Statistics is important. When samples are drawn from population N (µ, σ(2)) with a sample size of n, the distribution of the sample mean X̄ should be a normal distribution N (µ, σ(2)/n). Under the null hypothesis µ = µ0, the distribution of Statistics [Formula: see text] should be standardized as a normal distribution. When the variance of the population is not known, replacement with the sample variance s (2) is possible. In this case, the Statistics [Formula: see text] follows a t distribution (n-1 degrees of freedom). An independent-group t test can be carried out for a comparison of means between two independent groups, with a paired t test for paired data. As the t test is a Parametric test, samples should meet certain preconditions, such as normality, equal variances and independence.
Perussi, Livia Rodrigues - One of the best experts on this subject based on the ideXlab platform.
-
Avaliação da reparação de feridas cutâneas e ósseas realizadas por técnicas convencionais e com laser de Er, Cr:YSGG: estudo em ratos
Universidade Estadual Paulista (UNESP), 2010Co-Authors: Perussi, Livia RodriguesAbstract:O laser cirúrgico de Er,Cr:YSGG (érbio, impregnado por cromo: ítrio, scandio, gálio, garnet) tem sido utilizado no tratamento cirúrgico periodontal, embora poucos estudos histológicos sejam encontrados na literatura. O objetivo deste estudo foi comparar histologicamente a utilização do laser de Er-Cr: YSGG em tecido mole e ósseo em relação à técnicas cirúrgicas convencionais. Foram utilizados 20 ratos, divididos aleatoriamente em 3, 7, 15 e 30 dias, com 5 animais por período. Em cada animal foram realizados os procedimentos (P) nas regiões de dorso e calota, respectivamente: PI) Incisão no dorso com bisturi de lâmina no 15; PII) Incisão no dorso com Laser Er,Cr:YSGG (150mJ de energia, 2.0W de potência), PIII) Defeito ósseo na calota com fresa diamantada esférica em baixa rotação sob refrigeração; PIV) Defeito ósseo na calota com Laser Er,Cr:YSGG (150mJ de energia, 3.0W de potência). Os procedimentos foram randomicamente distribuídos entre lado direito e esquerdo de cada animal. A análise histológica foi avaliada de acordo com a intensidade da resposta inflamatória, presença de áreas necróticas e processo de regeneração tecidual. A análise histométrica dos defeitos ósseos foi confirmada pelo teste estatístico paramétrico ANOVA e complementado pelo teste Bonferroni. Para os procedimentos na região de dorso pode-se observar que nos períodos iniciais houve melhor resposta histológica para o PI, embora no período de 30 dias os procedimentos I e II apresentavam resultados semelhantes. A análise histométrica confirmou uma maior formação óssea na calota com o uso do laser de Er,Cr:YSGG aos 30 dias (79.96±10.30%) em relação ao procedimento com fresa aos 15 (62.05±4.84%) e 30 dias (58.23±9.99%). Na análise histológica em tecido duro, observou-se necrose óssea superficial...The Er,Cr:YSGG (Erbium, Chromium:Yttrium–Scandium–Gallium–Garnet) surgical laser has been demonstrated as an useful instrument on dentistry since its wavelength (2,78μm) is strongly absorbed by water and hydroxyapatite. The aim of this study was to histologically compare the in vivo healing process that occurs in soft and hard tissues incisions made by Er,Cr:YSGG laser or conventional periodontal surgeries techniques. Twenty rats were used and divided into 4 groups of 5 different periods. On each animal was performed the following procedures (P) in the soft and hard tissue: PI) dorsum incision with scalpel number 15; PII) dorsum irradiation with 2.0W Er,Cr:YSGG laser; PIII) skull defect made by a diamond spherical bur with water refrigeration; PIV) skull defect created by 3.0W Er,Cr:YSGG laser. The procedures were randomly distributed into right and left side of each animal. After sacrifice 3, 7, 15 and 30 days post-surgery the histological examinations were performed. Histometric analysis of the bone defects was confirmed by the Parametric Statistic test ANOVA and complemented by the Bonferroni test. In the first periods, the histological findings in the dorsum were better in the PI group, although by day 30 PI and PII showed similar healing response. The procedures in the skull revealed superficial necrosis in all periods in the PIV group, although this group had much more evidence of bone formation at day 30 than PIII. The histometric analysis also confirmed a better tissue response for laser application in the skull than conventional techniques, with 79.96±10.30% bone formation by day 30 with Er,Cr:YSGG laser and 62.05±4.84% by day 15 and 58.23±9.99% by day 30 with bur. The utilization of Er,Cr:YSGG laser in soft tissue caused a minimal delay in the healing initial periods but the... (Complete abstract, click electronic access below)Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq
Perussi, Livia Rodrigues [unesp] - One of the best experts on this subject based on the ideXlab platform.
-
Avaliação da reparação de feridas cutâneas e ósseas realizadas por técnicas convencionais e com laser de Er, Cr:YSGG: estudo em ratos
Universidade Estadual Paulista (UNESP), 2010Co-Authors: Perussi, Livia Rodrigues [unesp]Abstract:O laser cirúrgico de Er,Cr:YSGG (érbio, impregnado por cromo: ítrio, scandio, gálio, garnet) tem sido utilizado no tratamento cirúrgico periodontal, embora poucos estudos histológicos sejam encontrados na literatura. O objetivo deste estudo foi comparar histologicamente a utilização do laser de Er-Cr: YSGG em tecido mole e ósseo em relação à técnicas cirúrgicas convencionais. Foram utilizados 20 ratos, divididos aleatoriamente em 3, 7, 15 e 30 dias, com 5 animais por período. Em cada animal foram realizados os procedimentos (P) nas regiões de dorso e calota, respectivamente: PI) Incisão no dorso com bisturi de lâmina no 15; PII) Incisão no dorso com Laser Er,Cr:YSGG (150mJ de energia, 2.0W de potência), PIII) Defeito ósseo na calota com fresa diamantada esférica em baixa rotação sob refrigeração; PIV) Defeito ósseo na calota com Laser Er,Cr:YSGG (150mJ de energia, 3.0W de potência). Os procedimentos foram randomicamente distribuídos entre lado direito e esquerdo de cada animal. A análise histológica foi avaliada de acordo com a intensidade da resposta inflamatória, presença de áreas necróticas e processo de regeneração tecidual. A análise histométrica dos defeitos ósseos foi confirmada pelo teste estatístico paramétrico ANOVA e complementado pelo teste Bonferroni. Para os procedimentos na região de dorso pode-se observar que nos períodos iniciais houve melhor resposta histológica para o PI, embora no período de 30 dias os procedimentos I e II apresentavam resultados semelhantes. A análise histométrica confirmou uma maior formação óssea na calota com o uso do laser de Er,Cr:YSGG aos 30 dias (79.96±10.30%) em relação ao procedimento com fresa aos 15 (62.05±4.84%) e 30 dias (58.23±9.99%). Na análise histológica em tecido duro, observou-se necrose óssea superficial...The Er,Cr:YSGG (Erbium, Chromium:Yttrium–Scandium–Gallium–Garnet) surgical laser has been demonstrated as an useful instrument on dentistry since its wavelength (2,78μm) is strongly absorbed by water and hydroxyapatite. The aim of this study was to histologically compare the in vivo healing process that occurs in soft and hard tissues incisions made by Er,Cr:YSGG laser or conventional periodontal surgeries techniques. Twenty rats were used and divided into 4 groups of 5 different periods. On each animal was performed the following procedures (P) in the soft and hard tissue: PI) dorsum incision with scalpel number 15; PII) dorsum irradiation with 2.0W Er,Cr:YSGG laser; PIII) skull defect made by a diamond spherical bur with water refrigeration; PIV) skull defect created by 3.0W Er,Cr:YSGG laser. The procedures were randomly distributed into right and left side of each animal. After sacrifice 3, 7, 15 and 30 days post-surgery the histological examinations were performed. Histometric analysis of the bone defects was confirmed by the Parametric Statistic test ANOVA and complemented by the Bonferroni test. In the first periods, the histological findings in the dorsum were better in the PI group, although by day 30 PI and PII showed similar healing response. The procedures in the skull revealed superficial necrosis in all periods in the PIV group, although this group had much more evidence of bone formation at day 30 than PIII. The histometric analysis also confirmed a better tissue response for laser application in the skull than conventional techniques, with 79.96±10.30% bone formation by day 30 with Er,Cr:YSGG laser and 62.05±4.84% by day 15 and 58.23±9.99% by day 30 with bur. The utilization of Er,Cr:YSGG laser in soft tissue caused a minimal delay in the healing initial periods but the... (Complete abstract, click electronic access below
Cardona García Octavio - One of the best experts on this subject based on the ideXlab platform.
-
Identification and Modelling of Ultrafast Extreme Events in Financial Markets: Mini Flash Crashes = Identificación y modelado de eventos extremos en los mercados financieros: Mini Flash Crashes
2020Co-Authors: Cardona García OctavioAbstract:Mini Flash Crashes (MFC) or Ultrafast Extreme Events (UEE), are sudden and substantial movements in security prices with an equally fast reversion to their previous level. A MFC could last minutes, seconds, or even milliseconds. These kind of market failures have implications on its efficiency and increase the risk of larger drop downs. Since this is a relative new phenomenon in financial markets, some researchers have relied on ad-hoc procedures to identify these crashes. Aiming to find a more robust approach, this study consists in using the drift burst hypothesis jointly with a weighted on a grid logit model to identify and predict the probability of occurrence of Flash Crashes. The first approach detects short-lived explosive trends that cannot be explained by traditional volatility models. Each of these bursts is then labeled as a UEE. The Statistic relies on a non-Parametric estimator of the drift with a kernel window of five minutes. In contrast, the logit model solves an unbalanced learning problem where the minority class (UEE) represents only 1% of the entire sample. The performance of the classifier is compared against other techniques like the Synthetic Minority Oversampling Technique (SMOTE) and rebalanced classes, but also to other classifiers. In general terms, the non-Parametric Statistic identified more than 200 episodes in each asset composing the sample —SPDR Dow Jones Industrial Average ETF, SPDR S&P 500 ETF, and Apple— during two of the more remarkable Flash Crashes in history. On the other hand, the weighted logit model outperformed the other variations with an area under de curve (ROC-AUC) of 0.927 and accuracy of 0.895. Regarding policy implications, liquidity provision has significant effects on the probability of UEE. Reducing the speed of the transactions when the market is illiquid could be a better option than halting the market by using circuit brakers. Mini Flash Crashes (MFC) o Eventos Ultra rápidos y Extremos (UEE), son movimientos substanciales y repentinos del precio de un activo financiero con una rápida reversión a su nivel previo. Un MFC puede durar minutos, segundos, o incluso milisegundos. Este tipo de fallas en el mercado tiene implicaciones sobre su eficiencia e incrementa el riesgo de desplomes mucho más grandes. Dado que es un tema relativamente nuevo, algunos autores han tratado de identificar estos eventos utilizando metodologías ad hoc. Con el propósito de encontrar un procedimiento más robusto, este estudio consiste en utilizar la hipótesis “drift-burst” juntamente con un modelo logit ponderado, para identificar y predecir la probabilidad de que un Flash Crash ocurra. La primera metodología detecta tendencias explosivas cortas que no pueden explicarse por los modelos tradicionales de volatilidad, paso seguido cada una de estas ocurrencias se clasifica como un UEE. Este estadístico está basado en un estimador no paramétrico de la deriva con un kernel de cinco minutos. Análogamente, el modelo logit resuelve un problema de clasificación desbalanceado, en donde la categoría minoritaria (UEE) representa sólo el 1% de la muestra. El rendimiento de este modelo se compara contra otras técnicas y clasificadores. En términos generales, el estadístico no paramétrico identificó más de 200 ocurrencias en cada uno de los activos que componen la muestra —SPDR Dow Jones Industrial Average ETF— durante dos de los Flash Crashes más importantes de la historia. Por otro lado, el modelo logit ponderado supera a sus contrapartes con un área bajo la curva (ROC-AUC) de 0.927 y una precisión de 0.895. Con respecto a sus implicaciones en política, la provisión de liquides tiene efectos significativos en la probabilidad de ocurrencia de UEE. Reducir la velocidad de las transacciones cuando el mercado está ilíquido, puede ser una mejor opción que detener el mercado mediante “circuit breakers”
Wang Ding - One of the best experts on this subject based on the ideXlab platform.
-
Parametric Statistic and certification of physical stochastic model of seismic ground motion for engineering purposes
Journal of Earthquake Engineering and Engineering Vibration, 2013Co-Authors: Wang DingAbstract:Applying the PEER NGA Database,the Parametric probability distributions of the random function model of seismic ground motions are presented.According to the site classification in GB50011-2010,4438 ground motion records are classified.The sample values of the stochastic parameters are identified on different site classes using the system identification method.Based on the Parametric identification results,the Statistic analysis of the stochastic parameters in the random function model of seismic ground motions is made,and the probability density functions of the stochastic parameters are obtained.Selecting the representative point sets in the probability space of the random function parameters,the sample sets of the seismic ground motions are generated with the application of the superposition method of wave groups and the response spectrum sample sets are calculated.The Statistical feature comparisons of response spectra between the realistic ground motion records and the synthesized ground motion samples show the validity of the random function model and the Statistic results of the physical random parameters.