• Communications for Statistical Applications and Methods
• /
• v.13 no.2
• /
• pp.219-231
• /
• 2006

When the correlation of two random variables is weak, the value of one variable can not be used effectively to predict the other. Even when most of the values are overlapped, it is difficult to find a linear relationship. In this paper, we propose two graphical methods of representing the measures of correlation and independence between two random variables. The first method is used to represent their degree of correlation, and the other is used to represent their independence. Both of these methods are based on the cumulative distribution functions defined in this work.

• Journal of the Korean Statistical Society
• /
• v.19 no.2
• /
• pp.105-112
• /
• 1990

Let (X, Y) be a pair random variables and let f denote the regression function of the response Y on the measurement variable X. Let K(f) denote a derivative of f. The least squares method is used to obtain a tensor spline estimator $\hat{f}$ of f based on a random sample of size n from the distribution of (X, Y). Under some mild conditions, it is shown that $K(\hat{f})$ achieves the optimal rate of convergence for the estimation of K(f) in $L_2$ and $L_{\infty}$ norms.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1997.10a
• /
• pp.345-351
• /
• 1997

The motion of an aircraft landing gear over rough runway at variable speed is nonstationary. hi this paper, a method for the computation of nonstationary response variance is presented which uses a state space form for the combination of landing gear and runway excitation. The dynamic characteristics of the landing gear under nonstationazy random excitations has also been analyzed using the proposed method. The formulation is for linear systems of arbitrary order and allows any deterministic velocity history.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.1
• /
• pp.81-98
• /
• 2013

Since Probabilistic Latent Semantic Analysis (PLSA) and Latent Dirichlet Allocation (LDA) were introduced, many revised or extended topic models have appeared. Due to the intractable likelihood of these models, training any topic model requires to use some approximation algorithm such as variational approximation, Laplace approximation, or Markov chain Monte Carlo (MCMC). Although these approximation algorithms perform well, training a topic model is still computationally expensive given the large amount of data it requires. In this paper, we propose a new method, called non-simultaneous sampling deactivation, for efficient approximation of parameters in a topic model. While each random variable is normally sampled or obtained by a single predefined burn-in period in the traditional approximation algorithms, our new method is based on the observation that the random variable nodes in one topic model have all different periods of convergence. During the iterative approximation process, the proposed method allows each random variable node to be terminated or deactivated when it is converged. Therefore, compared to the traditional approximation ways in which usually every node is deactivated concurrently, the proposed method achieves the inference efficiency in terms of time and memory. We do not propose a new approximation algorithm, but a new process applicable to the existing approximation algorithms. Through experiments, we show the time and memory efficiency of the method, and discuss about the tradeoff between the efficiency of the approximation process and the parameter consistency.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.2
• /
• pp.676-682
• /
• 2018

As the IoT world including WSNs develops, the number of sensor systems that sense information according to the environment based on the principle of IoT is increasing. In order to perform security for each sensor system in such a complicated environment, the security modules must be varied. These problems make hardware/software implementation difficult when considering the system efficiency and hacking/cracking. Therefore, to solve this problem, this paper proposes a pseudorandom number generator (FLT: Pseudo-random Number Generator with Fixed Length Tap unrelated to the variable sensing nodes) with a fixed-length tap that generates a pseudorandom number with a constant period, irrespective of the number of sensing nodes, and has the purpose of detecting anomalies. The proposed FLT-LFSR architecture allows the security level and overall data formatting to be kept constant for hardware/software implementations in an IoT environment. Therefore, the proposed FLT-LFSR architecture emphasizes the scalability of the network, regardless of the ease of implementation of the sensor system and the number of sensing nodes.

• Journal of Biomedical Engineering Research
• /
• v.33 no.3
• /
• pp.148-154
• /
• 2012

Hemorrhagic shock is a primary cause of deaths resulting from injury in the world. Although many studies have tried to diagnose accurately hemorrhagic shock in the early stage, such attempts were not successful due to compensatory mechanisms of humans. The objective of this study was to construct a survival prediction model of rats in acute hemorrhagic shock using a random forest (RF) model. Heart rate (HR), mean arterial pressure (MAP), respiration rate (RR), lactate concentration (LC), and peripheral perfusion (PP) measured in rats were used as input variables for the RF model and its performance was compared with that of a logistic regression (LR) model. Before constructing the models, we performed 5-fold cross validation for RF variable selection, and forward stepwise variable selection for the LR model to examine which variables were important for the models. For the LR model, sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (ROC-AUC) were 0.83, 0.95, 0.88, and 0.96, respectively. For the RF models, sensitivity, specificity, accuracy, and AUC were 0.97, 0.95, 0.96, and 0.99, respectively. In conclusion, the RF model was superior to the LR model for survival prediction in the rat model.

• The Journal of the Korea Contents Association
• /
• v.6 no.5
• /
• pp.29-34
• /
• 2006

For the almost certainly convergent series $S_n=\sum_{i=1}^nV-i$ of independent random elements in Banach spaces, by investigating tail series laws of large numbers, the rate of convergence of the series $S_n$ to a random variable s is studied in this paper. More specifically, by studying the duality between the limiting behavior of the tail series $T_n=S-S_{n-1}=\sum_{i=n}^{\infty}V-i$ of random variables and that of Banach space valued random elements, an alternative way of proving a result of the previous work, which establishes the equivalence between the tail series weak law of large numbers and a limit law, is provided in a Banach space setting.

• Structural Engineering and Mechanics
• /
• v.91 no.1
• /
• pp.25-38
• /
• 2024

After the mainshock, whether the train can be allowed to pass the bridges plays an important role in ensuring the transport of supplies and rescue works for example, in the "12 May" earthquake in China, after evaluation, the bridge was still used for transportation in rescue at a very slow speed, engineers usually evaluate whether the train can pass the bridge safely based on the experience, lacks sufficient calculation basis and does not fully consider the risk caused by aftershocks. To address this issue, this paper comprehensively considers the randomness of track irregularity, the randomness of aftershock intensity and other multiple random sources in train-bridge interaction system (TBIS). The sensitivity of train to various random parameters after earthquake is analyzed from the perspective of probability, the most sensitive random variable in this paper is PGA of aftershocks, both for bridge and trailer car, With the increase of epicentral distance, the sensitivity of PGA will decrease, and correspondingly, for trailer car, the sensitivity of other random variables will increase, research in this paper provides a basis for the subsequent random analysis of post-earthquake driving safety.

• The Journal of Fisheries Business Administration
• /
• v.47 no.2
• /
• pp.15-32
• /
• 2016

The Association of Southeast Asian Nations(ASEAN) has been the most essential organization in Asia. In spite of the world economic crisis, Southeast Asian countries have shown fast economic growth since 2000, and they have been actively expanding investments and trades especially with major countries. Research on competitiveness in ASEAN market has spawned an increasingly large literature, but empirical research on the determinants of Korea's export to ASEAN is limited. The purpose of this study is to draw out the determinant of Korean fisheries export to ASEAN by carrying out a panel analysis. For achieving such a purpose, pooled OLS, Hausman Test, Fixed Effect, Random Effect are performed. The last 20 years' data over the period of 1995 to 2014 concentrated on the ASEAN 6 countries such as Indonesia, Malaysia, Philippine, Singapore, Thailand, Vietnam is used in this study. Amount of aquatic products export to ASEAN is used as the dependent variable; real exchange rate, real GDP, relative price level and GDP per capita are used as the explanatory variables and FTA as dummy variable. Empirical results show that fixed-effect analysis is the best model among all the models. As the fixed effect model shows, real exchange rate, real GDP, GDP per capita and dummy variable(FTA) play positive and statistically significant roles in fisheries export to ASEAN, while price variable plays a negative and statistically significant role to the dependent variable.

• International Area Studies Review
• /
• v.15 no.2
• /
• pp.151-173
• /
• 2011

The purpose of this paper is to examine the impact of production factors on economic growth in China during 1979~2008. In order to grasp the determinant and contribution analysis, we take fixed effect model and random effect model and Hausman test to choice model. The results show that the finance variable (rsav) and SOC variable (rsoc) have negative effects on the economic growth in the long run except some models. But unimproved raw labor variable (rlab), physical capital variable (rcap) and education variable (redu) shows strongly positive effect for the same time. We found the meaning of coefficients of growth factors. relative contribution of each input to per-capita growth in China. The direct elative contribution of physical investment to per-capita growth gives 35.9 percent in total model (TMO) and unimproved raw labor contributes only 4.7 percent. In all modes, physical investment (rcap) was the most important contributor of predicted growth in China economy.