• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.524-524
• /
• 2000

We have been developing a method to build one-step-ahead prediction models for time series using genetic programming (GP). Our model building method consists of two stages. In the first stage, functional forms of the models are inherited from their parent models through crossover operation of GP. In the second stage, the parameters of the newborn model arc optimized based on an iterative method just like the back propagation. The proposed method has been applied to various kinds of time series problems. An application to the seismic ground motion was presented in the KACC'99, and since then the method has been improved in many aspects, for example, additions of new node functions, improvements of the node functions, and new exploitations of many kinds of mutation operators. The new ideas and trials enhance the ability to generate effective and complicated models and reduce CPU time. Today, we will present a couple of financial applications, espc:cially focusing on gold price prediction in Tokyo market.

• Communications for Statistical Applications and Methods
• /
• v.21 no.2
• /
• pp.161-168
• /
• 2014

In mixture experiments with process variables, we consider the case that some of process variables are either uncontrollable or hard to control, which are called noise variables. Given the such mixture experimental data with process variables, first we study how to search for candidate models. Good candidate models are screened by the sequential variables selection method and checking the residual plots for the validity of the model assumption. Two methods, which use numerical optimization methods proposed by Derringer and Suich (1980) and minimization of the weighted expected loss, are proposed to find a cost-effective robust optimal condition in which the performance of the mean as well as the variance of the response for each of the candidate models is well-behaved under the cost restriction of the mixture. The proposed methods are illustrated with the well known fish patties texture example described by Cornell (2002).

• Food Science and Biotechnology
• /
• v.18 no.5
• /
• pp.1243-1252
• /
• 2009

User-friendly software was developed to determine the shelf-life of perishable Korean seasoned side dishes in real time based on growth models of spoilage and pathogenic microorganisms. In the program algorithm, the primary spoilage and fastest-growing pathogenic organisms are selected according to the product characteristics, and their growth is simulated based on the previously monitored or recorded temperature history. To predict the growth of spoilage organisms with confidence limits, kinetic models for aerobic bacteria or molds/yeasts from published works are used. Growth models of pathogenic bacteria were obtained from the literature or derived with regression of their growth rate data estimated from established software packages. These models are also used to check whether the risk of pathogenic bacterial growth exceeds that of food spoilage organisms. Many example simulations showed that the shelf-lives of the examined foods are predominantly limited by the growth of spoilage organism rather than by pathogenic bacterial growth.

• Journal of the Korea Safety Management & Science
• /
• v.9 no.1
• /
• pp.145-156
• /
• 2007

This paper is to propose model classification and evaluation of measurement uncertainty. In order to obtain type A and B uncertainty, variety of measurement mathematical models are illustrated by example. The four steps to evaluate expanded uncertainty are indicated as following; First, to get type A standard uncertainty, measurement mathematical models of single, double, multiple, design of experiment and serial autocorrelation are shown. Second, to solve type B standard uncertainty measurement mathematical models of empirical probability distributions and multivariate are presented. Third, type A and B combined uncertainty, considering sensitivity coefficient, linearity and correlation are discussed. Lastly, expanded uncertainty, considering degree of freedom for type A, B uncertainty and coverage factor are presented with uncertainty budget. SPC control chart to control expanded uncertainty is shown.

• East Asian mathematical journal
• /
• v.16 no.2
• /
• pp.251-266
• /
• 2000

The problem of solution uniqueness to the task of determining unknown parameters of mathematical models from input-output observations is studied. This problem is known as structural identifiability problem. We offer a new approach for testing structural identifiability of linear state space models. The approach compares favorably with numerous methods proposed by other authors for two main reasons. First, it is formulated in obvious mathematical form. Secondly, the method does not involve unfeasible symbolic computations and thus allows to test identifiability of large-scale models. In case of non-identifiability, when there is a set of solutions to the task, we offer a method of computing functions of the unknown parameters which can be determined uniquely from input-output observations and later used as new parameters of the model. Such functions are called parametric functions capable of estimation. To develop the method of computation of these functions we use Lie group transformation theory. Illustrative example is given to demonstrate applicability of presented methods.

• Communications for Statistical Applications and Methods
• /
• v.28 no.4
• /
• pp.329-338
• /
• 2021

Deep learning methods have been developed, used in various fields, and they have shown outstanding performances in many cases. Many studies predicted a daily stock return, a classic example of time-series data, using deep learning methods. We also tried to apply deep learning methods to Korea's stock market data. We used Korea's stock market index (KOSPI) and several individual stocks to forecast daily returns and directions. We compared several deep learning models with other machine learning methods, including random forest and XGBoost. In regression, long short term memory (LSTM) and gated recurrent unit (GRU) models are better than other prediction models. For the classification applications, there is no clear winner. However, even the best deep learning models cannot predict significantly better than the simple base model. We believe that it is challenging to predict daily stock return data even if we use the latest deep learning methods.

• Communications for Statistical Applications and Methods
• /
• v.30 no.1
• /
• pp.49-63
• /
• 2023

A lot of studies on the summary measures of predictive strength of categorical response models consider the likelihood ratio index (LRI), also known as the McFadden-R², a better option than many other measures. We propose a simple modification of the LRI that adjusts for the effect of the number of response categories on the measure and that also rescales its values, mimicking an underlying latent measure. The modified measure is applicable to both binary and ordinal response models fitted by maximum likelihood. Results from simulation studies and a real data example on the olfactory perception of boar taint show that the proposed measure outperforms most of the widely used goodness-of-fit measures for binary and ordinal models. The proposed R² interestingly proves quite invariant to an increasing number of response categories of an ordinal model.

• Bulletin of the Society of Naval Architects of Korea
• /
• v.16 no.2
• /
• pp.9-14
• /
• 1979

Havelock was considered the wave resistance of a post extending vertically downwards through the water from the surface, its section by a horizontal plane being the same at all depths and having its breath small compared with its length. This enables us to elucidate certain points of interest in ship resistance. However, the ship has not infinte draft. So, the problem which is investigated in detail in this paper is the wave resistance of a mathematical quadratic model in a uniform stream. The author wishes to study the effect of viriation of draft. The form of the water-plane is varied while keeping in length and the cross sectional area constant. As a numerical example, we calculated the wave resistance for mathematical quadratic ship models. The results are compared with a post having infinite depth. The results are as follows; The models with finer ends have smaller wave resistance up to $V/\sqrt{L}=1.1{\sim}1.2$ than its mate with blunter ends, but above this speed the models with blunter ends have less wave resistance. According to the decrease of draft, the wave resistance gap between the models with blunter ends the models with finer ends decrease at high speed. In this case of T/L=0.025, the models with finer ends have less wave resistance than the models with blunter ends at high speed.

• Research in Mathematical Education
• /
• v.26 no.3
• /
• pp.167-202
• /
• 2023

A central problem of mathematics teaching worldwide is probably the insufficient adaptive handling of tasks-especially in computational practice phases and modeling tasks. All students in a classroom must often work on the same tasks. In the process, the high-achieving students are often underchallenged, and the low-achieving ones are overchallenged. This publication uses different modeling of the tennis serve as an example to show a possible solution to the problem and develops and discusses one adaptive task each for middle school, high school, and college using three mathematical models of the tennis serve each time. From model to model within the task, the complexity of the modeling increases, the mathematical or physical demands on the students increase, and the new modeling leads to more realistic results. The proposed models offer the possibility to address heterogeneous learning groups by their arrangement in the surface structure of the so-called parallel adaptive task and to stimulate adaptive mathematics teaching on the instructional topic of mathematical modeling. Models A through C are suitable for middle school instruction, models C through E for high school, and models E through G for college. The models are classified in the specific modeling cycle and its extension by a digital tool model, and individual modeling steps are explained. The advantages of the presented models regarding teaching and learning mathematical modeling are elaborated. In addition, we report our first teaching experiences with the developed parallel adaptive tasks.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.10 no.1
• /
• pp.1-8
• /
• 1985

By introducing the concept of reliability of parallel systems the set covering. A branch-and-bound algorithm is developed and illustrated by a numerical example. The procedure has been coded and its computational efficiency is studied.