• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.08a
• /
• pp.470-474
• /
• 1998

In this paper, to estimate the time-varying parameters of speech signal, we use the robust sequential estimator based on t-distribution and, for time-varying signal, introduce the forgetting factor. By using the RSE based on t-distribution with small degree of freedom, we can alleviate efficiently the effects of outliers to obtain the better performance of parameter estimation. Moreover, by the forgetting factor, the proposed algorithm can estimate the accurate parameters under the rapid variation of speech signal.

• International Area Studies Review
• /
• v.12 no.2
• /
• pp.3-26
• /
• 2008

This article investigates the usefulness of the skewed Student-t distribution in modeling the long memory volatility property that might be present in the daily returns of two Australian financial series; the ASX200 stock index and AUD/USD exchange rate. For this purpose we assess the performance of FIGARCH and FIAPARCH Value-at-Risk (VaR) models based on the normal, Student-t, and skewed Student-t distribution innovations. Our results support the argument that the skewed Student-t distribution models produce more accurate VaR estimates of Australian financial markets than the normal and Student-t distribution models. Thus, consideration of skewness and excess kurtosis in asset return distributions provides appropriate criteria for model selection in the context of long memory volatility models in Australian stock and foreign exchange markets.

• Journal of Power System Engineering
• /
• v.17 no.4
• /
• pp.86-93
• /
• 2013

This paper deals with the effects of driving force and material properties on statistical distribution of fatigue crack growth rate (FCGR) for the friction stir welded joints of Al 7075-T651 aluminum plate. In this work, the statistical probability distribution of fatigue crack growth rate was analyzed by using our previous constant stress intensity factor range controlled fatigue crack growth test data. As far as this study are concerned, the statistical probability distribution of fatigue crack growth rate for the friction stir welded (FSWed) joints was found to evaluate the variability of fatigue crack growth rate for base metal (BM), heat affected zone (HAZ) and weld metal (WM) specimens. The probability distribution of fatigue crack growth rate for FSWed joints was found to follow well log-normal distribution. The shape parameter of BM and HAZ was decreased with increasing the driving force, however, the shape parameter of WM was decreased and increased with increasing the driving force. The scale parameter of BM, HAZ and WM was increased with the driving force.

• Communications for Statistical Applications and Methods
• /
• v.15 no.5
• /
• pp.667-674
• /
• 2008

We develop new family of the t distributions for modeling semicircular data. It has convenient mathematical features. It is extended to the l-axial t distribution and a generalized semicircular t distribution.

• Korean Journal of Ecology and Environment
• /
• v.52 no.1
• /
• pp.21-27
• /
• 2019

Analysis of water quality distribution is very important for river water quality management. Recently, various studies have been conducted on the analysis of water quality distribution according to reduction methods of nonpoint pollutant. The objective of this study was to select the probability distributions of water quality constituents (T-N, T-P, COD, SS) according to the farming forms (control, slow release fertilizer, water depth control) during non-storm period in the paddy fields. The field monitoring was conducted monitoring site located in Baeksan-myun, Buan-gun, Jeollabuk-do, Korea during non-storm period from May to September in 2016. Our results showed that there were no differences in water quality among three different farming forms, except for SS of control and water depth control. K-S method was used to analyzed the probability distributions of T-N, T-P, COD and SS concentrations discharged from paddy fields. As a results of the fitness analysis, T-N was not suitable for the normal probability distribution in the slow release fertilizer treatment, and the log-normal probability distribution was not suitable for the T-P in control treatment. The gamma probability distribution showed that T-N and T-P in control and slow release fertilizer treatment were not suitable. The Weibull probability distribution was found to be suitable for all water quality constituents of control, slow release fertilizer, and water depth control treatments. However, our results presented some differences from previous studies. Therefore, it is necessary to analyze the characteristics of pollutants flowing out in difference periods according to various farming types. The result of this study can help to understand the water quality characteristics of the river.

• Journal of Convergence for Information Technology
• /
• v.10 no.11
• /
• pp.23-29
• /
• 2020

As IoT devices increase exponentially, minimizing MIMO interference and increasing transmission capacity for sending and receiving IoT information through multiple antennas remain the biggest issues. In this paper, secret key-level distribution mechanism using deep learning is proposed to minimize MIMO-based IoT communication noise. The proposed mechanism minimizes resource loss during transmission and reception process by dispersing IoT information sent and received through multiple antennas in batches using deep learning. In addition, the proposed mechanism applied a multidimensional key distribution processing process to maximize capacity through multiple antenna multiple stream transmission at base stations without direct interference between the APs. In addition, the proposed mechanism synchronizes IoT information by deep learning the frequency of use of secret keys according to the number of IoT information by applying the method of distributing secret keys in dimension according to the number of frequency channels of IoT information in order to make the most of the multiple antenna technology.

• Journal of the Korean Statistical Society
• /
• v.35 no.4
• /
• pp.419-439
• /
• 2006

The article develops a new class of distributions by introducing a nonnegative perturbing function to $t_\nu$ distribution having location and scale parameters. The class is obtained by using transformations and conditioning. The class strictly includes $t_\nu$ and $skew-t_\nu$ distributions. It provides yet other models useful for selection modeling and robustness analysis. Analytic forms of the densities are obtained and distributional properties are studied. These developments are followed by an easy method for estimating the distribution by using Markov chain Monte Carlo. It is shown that the method is straightforward to specify distribution ally and to implement computationally, with output readily adopted for constructing required criterion. The method is illustrated by using a simulation study.

• Bulletin of the Korean Mathematical Society
• /
• v.34 no.2
• /
• pp.287-293
• /
• 1997

(t); t \geq 0} on R^1$ satisfying the following stochastic differential equation.

• Proceedings of the KSRS Conference
• /
• v.1
• /
• pp.253-256
• /
• 2006

When the background tropospheric ozone column over the Pacific Ocean is subtracted from the latitudinal total ozone distribution, the results show remarkable agreement with the latitudinal stratospheric ozone distribution using the CCD. The latitudinal tropospheric ozone distribution using the CCD method, with a persistent maximum over the southern tropical Atlantic, is also seen in the latitudinal tropospheric ozone distribution using the T-P method. It suggests that the CCD method can be replaced by the simple T-P method. However, the tropical Atlantic paradox exists in the results of both the CCD and T-P methods during the northern burning season. In order to investigate this paradox, we compare the latitudinal ozone distributions using the CCD and T-P methods by using the SAGE measurements (e.g. TSA method) and the SHADOZ ozonesoundings (e.g. T-S method) assuming zonally invariant stratospheric ozone, which is the same assumption as of the CCD method. During the northern burning season, the latitudinal distributions in the tropospheric ozone derived from the T-SA and T-S methods show higher tropospheric ozone over the northern tropical Atlantic than the southern Atlantic due to a stronger gradient in stratospheric ozone relative to that from the CCD and T-P methods. This indicates that the latitudinal tropospheric ozone distribution can be changed depending on the data that is used to determine the latitudinal stratospheric ozone distribution. Therefore, there is a possibility that the north-south gradient in stratospheric ozone over the Atlantic can be a solution of the paradox.

• Communications for Statistical Applications and Methods
• /
• v.4 no.1
• /
• pp.281-292
• /
• 1997

In this paper we are concered with the counting processes with intersity function $g_n(t)$, where $g_n(t)$ not only depends on t but n. It is shown that under certain conditions the number of events in [0, t] follows a generalizes Poisson distribution. A counting process is also provided such that $g_i(t)$$\neq$$g_i(t)$ for i$\neq$j and the number of events in [0, t] has a transformed geometric distribution.