• International Journal of Contents
• /
• v.8 no.4
• /
• pp.30-35
• /
• 2012

Error surfaces provide us with very important information for training of feed-forward neural networks (FNNs). In this paper, we draw the contour plots of various error or objective functions for training of FNNs. Firstly, when applying FNNs to classifications, the weakness of mean-squared error is explained with the viewpoint of error contour plot. And the classification figure of merit, mean log-square error, cross-entropy error, and n-th order extension of cross-entropy error objective functions are considered for the contour plots. Also, the recently proposed target node method is explained with the viewpoint of contour plot. Based on the contour plots, we can explain characteristics of various error or objective functions when training of FNNs proceeds.

• Journal of Korean Society of Forest Science
• /
• v.80 no.4
• /
• pp.379-382
• /
• 1991

Tree height at age 10 was used to estimate the statistical efficiencies of sampling size in the progeny test of Pinus koraiensis S. et Z. Experimental design was RCB design which consists of 25 half-sib families in each of three blocks. The number of families and blocks were fixed, therefore, the number of trees sampled per plot was the only factor that influences the environmental portion of the family mean height. Coefficient of variation, the estimate of the standard error of the family mean height, decreased with increase of sampling plot size, and became stable from 4-tree plot sampling (6.97%). The experimental error was significant from 7-tree sampling plot size. Nonlinear relationship (${\hat{Y}}=10.425e-^{0.073x}$ ; $R^2$=0.840) was found between the sampling plot size and the standard error of family mean height.

• Journal of the Korean Geophysical Society
• /
• v.5 no.2
• /
• pp.131-142
• /
• 2002

Geomagnetic transfer function is generally estimated by choosing transfer to minimize the square sum of differences between observed values. If the error structure sccords to the Gaussian distribution, standard least square(LS) can be the estimation. However, for non-Gaussian error distribution, the LS estimation can be severely biased and distorted. In this paper, the Gaussian error assumption was tested by Q-Q(Quantile-Quantile) plot which provided information of real error structure. Therefore, robust estimation such as regression M-estimate that does not allow a few bad points to dominate the estimate was applied for error structure with non-Gaussian distribution. The results indicate that the performance of robust estimation is similar to the one of LS estimation for Gaussian error distribution, whereas the robust estimation yields more reliable and smooth transfer function estimates than standard LS for non-Gaussian error distribution.

• Communications for Statistical Applications and Methods
• /
• v.4 no.1
• /
• pp.139-149
• /
• 1997

For split plot designs exact univariate F tests of the within-subjects main effect are based on the assumption of multisample sphericity. Type I error rates and power are reported for the F test and two tests designed for use when multisample sphericity is violated: the $\widetilde{\xi}$-adjusted test and the Corrected Improved General Approximation(CIGA) test.The results indicate that even though the F test and the $\widetilde{\xi}$-adjusted test have better power than the CIGA test in some conditions, the F test and the $\widetilde{\xi}$-adjusted test do not control Type I error rates when the design is unbalanced and the F test dose not have a good control of Type I error rates when sphericity assumption is severely violated.

• The Korean Journal of Applied Statistics
• /
• v.27 no.2
• /
• pp.227-237
• /
• 2014

We use the regression model with the indicator variables in the case that we use qualitative variables as some predictor variables in regression analysis. We use the ANCOVA(Analysis of Covariance) model when comparing the response variable among groups while statistically controlling for variation in the response variable caused by a variation in the covariate. DD-plot can be used as a graphical exploratory data analysis tool before the confirmatory data analysis. With the DD-plot, we can discriminate the difference of groups in the regression model with the indicator variables or the ANCOVA model at a glance. Making DD-plot does not demand the statistical model assumption about error terms in regression model. Several examples show the usefulness of DD-plots as a graphical exploratory data analysis tool for the regression analysis.

• Proceedings of the KSME Conference
• /
• 2004.04a
• /
• pp.780-785
• /
• 2004

The work presented here deals with controller design by graphical method based on proportional control ramp response. The design aims at the improvement of transient response, disturbance rejection capability, steady-state error reduction with stability preservation. The first step is to generate tracking-error curve with proportional control only and decide the added error signal shape on the error curve. The effectiveness of the proposed controller is confirmed through the simulation and experiment.

• Journal of the Korea Society for Simulation
• /
• v.29 no.2
• /
• pp.83-93
• /
• 2020

Bivariate measure of risk and error(BMORE) plot is originally designed to depict bivariate output data and related statistics obtained from a stochastic simulation such as sample mean, median, outliers, and a boundary of a certain percentile of simulation data. When compared to the static numbers, the plot has a big advantage in visualization that enables scholars and practitioners to understand the potential variability and risk in the simulation data. In this study, beyond just the construction of the plot to depict the variability of a certain system, we add a chance constraint to the plot and apply it for decision making such as checking the feasibility of systems, comparing performances of the systems on statistical background, and also analyzing the sensitivity of the problem parameters. In order to demonstrate an application of the plot, we employ an inventory management problem as an example. However, the techniques and algorithms suggested in this paper can be applied to any other problems comparing systems on bivariate performance measures with simulation/experiment results.

• Proceedings of the Safety Management and Science Conference
• /
• 2008.04a
• /
• pp.295-299
• /
• 2008

This paper discusses tests of factor effect or contrast by the use of saturated design $k^n$ factorial design. The nine nonparametric rank measures in normality test using normal probability pot are proposed. Length's PSE(Pseduo Standard Error) test [4] which relies on the concept of effect sparsity is also introduced and extended to the margin of error(ME) and Simultaneous margin of error(SME).

• Journal of Industrial Technology
• /
• v.14
• /
• pp.127-140
• /
• 1994

This study is concerned with the accuracy and the patterns when different methods was used in interpolation task. Although 3 methods employed the same modality for input (visual) and for output (manual responding), they differed in central processing, which method 1 is relatively more tendency of verbal processing, method 2 is realtively more tendency of spatial processing and method 3 needed a number of switching code (verbal/spatial) performing task. Split-plot design was adopted, which whole plot consisted of methods (3), orientations (horizon, vertical), base-line sizes (300, 500, 700 pixels) and split plot consisted of target locations (1-99). The results showed the anchor effect and the range effect. Method 2, method 3 and method 1 that order was better accuracy. ANOVA showed that the accuracy was significantly influenced by the method, the location of target, and its interactions ($method{\times}location$, $size{\times}location$). Analysis of error data, response time and frequency of under, just, over estimate indicated that a systematic error pattern was made in task and methods changed not only the performance but also the pattern. The results provided support for the importance of the multiple resources theory in accounting for S-C-R compatibility and task performance. They are discussed in terms of multiple resources theory and guidelines for system design is suggested by the S-C-R compatibility.

• 한국전산유체공학회:학술대회논문집
• /
• 2004.10a
• /
• pp.121-125
• /
• 2004

The preprocessor - solver - postprocessor software for 2D/Axisymmetric CSCM Upwind Flux Difference Splitting Navier-Stokes code has been developed for undergraduate educational purpose. This computational fluid dynamics (CFD) software allows students to setup, solve, visualize and control dynamically server for their own fluid problems via Internet. The preprocessor Is capable of generating geometry and grid, initial solution data and required solver control parameters. The postprocessor shows vector plot and contour plot with different options while residual plot shows root-mean-square (RMS) error history graphically and retrieves the data from solver interactively. Special feature of the preprocessor is grid generation part which is based on MFC/Visual C++ application and FORTRAN single block grid generator process. Many users can access solver via Internet from client computers and solve desired problems using locally installed pre- and postprocessor and remote powerful solver part.