• Communications for Statistical Applications and Methods
• /
• v.28 no.1
• /
• pp.39-57
• /
• 2021

Time series prediction is an area of great interest to many people. Algorithms for time series prediction are widely used in many fields such as stock price, temperature, energy and weather forecast; in addtion, classical models as well as recurrent neural networks (RNNs) have been actively developed. After introducing the attention mechanism to neural network models, many new models with improved performance have been developed; in addition, models using attention twice have also recently been proposed, resulting in further performance improvements. In this paper, we consider time series prediction by introducing attention twice to an RNN model. The proposed model is a method that introduces H-attention and T-attention for output value and time step information to select useful information. We conduct experiments on stock price, temperature and energy data and confirm that the proposed model outperforms existing models.

• Journal of the Korea Society for Simulation
• /
• v.17 no.3
• /
• pp.19-26
• /
• 2008

In this paper, we propose a dynamic prediction algorithm to predict the bond price using actual data set of treasure note (T-Note). The proposed algorithm is based on term structure model of the interest rates, which takes place in various financial modelling, such as the standard Gaussian Wiener process. To obtain cumulative distribution functions (CDFs) of actual data for the interest rate measurement used, we use the natural cubic spline (NCS) method, which is generally used as numerical methods for interpolation. Then we also use the random number generation scheme (RNGS) to calculate the pricing of bond through the obtained CDF. In empirical computer simulations, we show that the lower values of precision in the proposed prediction algorithm corresponds to sharper estimates. It is very reasonable on prediction.

• Atmosphere
• /
• v.33 no.4
• /
• pp.355-365
• /
• 2023

To proactively manage climate risk, near-term climate predictions on annual to decadal time scales are of great interest to various communities. This study evaluates the near-term climate prediction skills in East Asia with DePreSys4 retrospective decadal predictions. The model is initialized every November from 1960 to 2020, consisting of 61 initializations with ten ensemble members. The prediction skill is quantitatively evaluated using the deterministic and probabilistic metrics, particularly for annual mean near-surface temperature, land precipitation, and sea level pressure. The near-term climate predictions for May~September and November~March averages over the five years are also assessed. DePreSys4 successfully predicts the annual mean and the five-year mean near-surface temperatures in East Asia, as the long-term trend sourced from external radiative forcing is well reproduced. However, land precipitation predictions are statistically significant only in very limited sporadic regions. The sea level pressure predictions also show statistically significant skills only over the ocean due to the failure of predicting a long-term trend over the land.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.11
• /
• pp.127-135
• /
• 2020

This study aims to measure the effects of real interest rate (RIR), gross domestic savings (GDS), and net exports (EN) shocks on Indonesia's economic growth (EG). The focus on Indonesia is unique due to the abundant resources available in the nation, but they are unsuccessful in boosting economic growth. This study applied a quantitative method to comprehensively analyze the correlation between variables by employing Vector Autoregression Model (VAR) combined with Vector Error Correction Model (VECM). Various procedures are preformed: Augmented Dickey-Fuller test (ADF), Optimum Lag Test, Johansen Cointegration Test, Granger Causality Test, as well as Impulse Response Function (IRF) and Error Variance Decomposition Analysis (FEVD). The data were collected from the World Bank and the Asian Development Bank from 1986 to 2017. The findings of the study indicated that economic growth responded positively to real interest rate shocks, which implies that when the real interest rate experiences a shock (increase), the economy will be inclined to growth. While, economic growth responded negatively to gross domestic savings and net export shocks. Policymakers are expected to consider several matters, particularly the economic conditions at the time of formulating policy, so that the prediction effectiveness of a policy can be appropriately assessed.

• Journal of the Korean Statistical Society
• /
• v.3 no.2
• /
• pp.103-115
• /
• 1974

During the last few years there has been an interest in models for qualitative attributes, where complex signifies that affection may be caused in two or more ways [1-3]. These models have in common the prediction of variable recurrence risks among families with given parental phenotpes. Segregation analysis has covered only a few cases [4,5]. The present paper extends segregation analysis to three complex models under two mode of ascertainment.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.6 no.3
• /
• pp.86-102
• /
• 2003

Polynitrogen molecules are of great interest as potential high energy-density materials, and hence such structures of various isomers of nitrogen clusters have been calculated using molecular modeling techniques by the researchers from various sectors of scientific institutions. In this article, the predicted meta-stable structures of these hypothetical molecules have been thoroughly reviewed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.1
• /
• pp.33-46
• /
• 1996

Inthis paper, we investigate the compatible coding technique, which receives much interest ever since the introduction of HDTV. First, attempts have been made to analyze the theoretical transform coding gains for various hierarchical decomposition techniques, namely subband, pyramid and DCT-based decomposition techniques. It is shown that the spatical domain techniques proide higher transform coding gains than the DCT-based coding technique. Secondly, we compare the performance of these spatial domain techniques, in terms of the PSNR versus various rate allocations to each layer. Based on these analyses, it is believed that the pyramid decomposition is more appropriate for the compatible coding. Also in this paper, we propose a hybrid prediction pyramid coding technique, by combining the spatio-temporal prediction in MPEG-2[3] and the adaptive MC(Motion Compensation)[1]. In the proposed coding technigue, we also employ an adaptive DCT coefficient scanning technique to exploit the direction information of the 2nd-layer signal. Through computer simulations, the proposed hybrid prediction with adaptive scanning technuque shows the PSNR improvement, by about 0.46-1.78dB at low 1st-layer rate(about 0.1bpp) over the adaptive MC[1], and by about 0.33-0.63dB at high 1st-layer rate (about 0.32-0.43bpp) over the spatio-temporal prediction[3].

• IEIE Transactions on Smart Processing and Computing
• /
• v.1 no.2
• /
• pp.78-87
• /
• 2012

In this paper, we address the tracking problem caused by camera motion and rolling shutter effects associated with CMOS sensors in consumer handheld cameras, such as mobile cameras, digital cameras, and digital camcorders. A modified particle filtering method is proposed for simultaneously tracking objects and compensating for the effects of camera motion. The proposed method uses an elastic registration algorithm (ER) that considers the global affine motion as well as the brightness and contrast between images, assuming that camera motion results in an affine transform of the image between two successive frames. By assuming that the camera motion is modeled globally by an affine transform, only the global affine model instead of the local model was considered. Only the brightness parameter was used in intensity variation. The contrast parameters used in the original ER algorithm were ignored because the change in illumination is small enough between temporally adjacent frames. The proposed particle filtering consists of the following four steps: (i) prediction step, (ii) compensating prediction state error based on camera motion estimation, (iii) update step and (iv) re-sampling step. A larger number of particles are needed when camera motion generates a prediction state error of an object at the prediction step. The proposed method robustly tracks the object of interest by compensating for the prediction state error using the affine motion model estimated from ER. Experimental results show that the proposed method outperforms the conventional particle filter, and can track moving objects robustly in consumer handheld imaging devices.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.10a
• /
• pp.199-202
• /
• 1996

Although numerous studies demonstrate that one technique outperforms the others for a given data set, there is often no way to tell a priori which of these techniques will be most effective to solve a specific problem. Alternatively, it has been suggested that a better approach to classification problem might be to integrate several different forecasting techniques by combining their results. The issues of interest are how to integrate different modeling techniques to increase the prediction performance. This paper proposes the post-model integration method, which means integration is performed after individual techniques produce their own outputs, by finding the best combination of the results of each method. To get the optimal or near optimal combination of different prediction techniques. Genetic Algorithms (GAs) are applied, which are particularly suitable for multi-parameter optimization problems with an objective function subject to numerous hard and soft constraints. This study applied three individual classification techniques (Discriminant analysis, Logit and Neural Networks) as base models to the corporate failure prediction context. Results of composite prediction were compared to the individual models. Preliminary results suggests that the use of integrated methods will offer improved performance in business classification problems.

• Computers and Concrete
• /
• v.12 no.2
• /
• pp.187-209
• /
• 2013

This paper presents a study on the prediction of transfer length of 13 mm seven-wire prestressing steel strand in pretensioned prestressed concrete members with rectangular cross-section including several material properties and design and manufacture parameters. To this end, a carefully selected database consisting of 207 different cases coming from 18 different sources spanning a variety of practical transfer length prediction situations was compiled. 16 single input features and 5 combined input features are analyzed. A widely used feedforward neural regression model was considered as a representative of several machine learning methods that have already been used in the engineering field. Classical multiple linear regression was also considered in order to comparatively assess performance and robustness in this context. The results show that the implemented model has good prediction and generalization capacity when it is used on large input data sets of practical interest from the engineering point of view. In particular, a neural model is proposed -using only 4 hidden units and 10 input variables-which significantly reduces in 30% and 60% the errors in transfer length prediction when using standard linear regression or fixed formulas, respectively.