Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
Korean Journal of Applied Statistics
Journal Basic Information
Journal DOI :
The Korean Statistical Society
Editor in Chief :
Volume & Issues
Volume 5, Issue 2 - Sep 1992
Volume 5, Issue 1 - Mar 1992
Selecting the target year
Statistical interpolation of meteorological data
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 113~121
Statistical interpolation which uses past experience about the behaviour of the atmosphere(correlation function) to interpolate the observations irregularly distributed in space and time to regular grids is discussed. Correlation functions are computed in the Far East Asian region during the winter periods of December 1977 to February 1980. Results show from the computation of correlation functions that there exists a large difference in the autocorrelation functions by slowly decreasing geopotential height and temperature, and rapidly decreasing wind and mixing ratio with increasing data-correlation functions between geopotential height and wind are well corresponed to persistence of wintertime synoptic features.
Statistical nature of the dry and wet periods defined in the time series of annual precipitations (1771-1990) of seoul
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 123~137
We analyzed a time series composed of the annual precipitations of Seoul based on the measurements of a Korean raingage and a modern raingage. The precipitations measured with a Korean raingage for the period of 1771 to 1907 are followed by the precipitations with a modern raingage for the period of 1908 to 1990. The latter part of the time series of annual precipitations were obtained from a book for annual precipitations of Korea by Korea Meteorological Administration and the former from Wada's table 1 for monthly precipitations reproduced from the daily rainfall measurements by a Korean raingage for the period of the Yi Dynasty. In our analysis three different precipitation regimes clearly stand out of the entire period. In order to define objectively the period of each precipitation regime we made a time series of 9 year moving averages from the above time series. By taking into account the shapes of the moving average time series and by using a threshold value of annual precipitation 1050 mm, we defined three precipitation regimes of wet period 1(WP1), dry period (DP), and wet period 2 (WP2). The WP1 and WP2 show very similar characteristics in out statistical analyses. On the other hand, DP is very different from the two periods in many statistical aspects. The strong similarities of the WP1 and WP2 regimes in the magnitudes of statistical parameters and in the shapes of their power spectrum distribution are supporting very positively the soundness of precipitation amounts measured with a Korean raingage in spite of numerous conceivable errors which might have been introduced into measurements of precipitation due to changes of observation site and environment, the scale of units employed, and urbanization of Seoul, etc. However, the annual precipitation amounts are not enough to examine throughly the characteristic of precipitation variations during the two regimes. It is definitely necessarly to recover the daily amounts of precipitation, based on two or three times measurements of rainfall with a Korean raingage, scattered in various ancient documents such as the official diary of 'Seungjeong-weon'
Outlier detection and time series modelling in the stationary time series
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 139~156
Recently several authors have introduced iterative methods for detecting time series outliers. Most of these methods are developed under the assumption that an underlying outlier-free model is known or can be identified. Since outliers can distort model identification or even make it impossible, we propose procedure begins with a descriptive data analysis of a time series using distance measures between two observations. Properties of the proposed test statistic are presented. To distinguish the type of an outlier are used transfer function models. An empirical example is given to illustrate the time series modeling procedure.
A new two-state randomized response model
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 157~167
This paper presents a new two-stage randomized response model to protect greater privacy of respondents for the sensitive characters. The conditions when the proposed model will be more efficient than Warner model, Liu-Chow's multiple trial model and Mangat-Singh model have been obtained for the case when the respondents are truthful in their answer, and the efficiency of the proposed model is also compared with Warner model, Liu-Chow's multiple trial model and Mangat-Singh model.
Some orthogonal factorial row-column designs
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 169~179
It is shown that a structurally complete row-column design has orthogonal factorial structure if each of its component designs has orthogonal factorial structure. It implies that such designs are most easily constructed via the amalgamating of one-dimensional block designs which have orthogonal factorial structure. However, this does not always hold for structurally incomplete row-column designs. A structurally incomplete row-column design is derived from the design with adjusted orthogonality, by simply interchanging row and treatment numbers.
A study on selection of tensor spline models
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 181~192
We consider the estimation of the regression surface in generalized linear models based on tensor-product B-splines in a data-dependent way. Our approach is to use maximum likelihood method to estimate the regression function by a function from a space of tensor-product B-splines that have a finite number of knots and are linear in the tails. The knots are placed at selected order statistics of each coordinate of the sample data. The number of knots is determined by minimizing a variant of AIC. A numerical example is used to illustrate the performance of the tensor spline estimates.
On a bivariate step-stress life test
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 193~209
We consider a Step Life Testing which is deviced for a two-component serial system with the considerably long life time. In the modelling stage we discuss the bivariate exponential distribution suggested by Block and Basu as the bivariate survival function for the two-component system, and develope the cumulative exposure model introduced by Nelson so that it can be used under the bivariate function. We consider inference on the component life time when the components are at work in the system by combining the information from system life test and that from the component tests carried out separately under the controlled environment. In data analysis, maximum likelihood estimators are discussed with the initial value obtained by an weighted least square method. Finally we discuss the optimal time for changing the stress in the simple step stress life testing.
A study on the relation between dissimilarity and hierarchical agglomerative in clust analysis
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 211~227
In this paper we consider the definition and mathematical properties of similarity or dissimilarity which have often used in clust analysis, and we apply a hierarchical agglomerative cluster algorithm to a dissimilarity metrx generated by these distance. Here we investigate the effect of relation between distance function and cluster algorithm on the retrieval ability of natural clusters. We present an empirical results for qualitative data as well as quantitative data.
On analysis of row-column designs
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 229~242
Bradley and Stewart(1991) considered a large class of experimental designs as multidimensional block designs(MBD's). The simplest MBD could be considered to be a row-column design(RCD). They presented the intrablock analysis of variance for a general row-column design. In this article, a generalized least squares solution for Bradley & Stewart's example is considered. In this case, the assumption is that row and column effects are random. This is an application of revised Paik(1990a,1990b)'s method. The Appendix is devoted to that revised method.
Study on the analysis of disproportionate data and hypothesis testing
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 243~254
In the present study two sets of unbalanced two-way cross-classification data with and without empty cell(s) were used to evaluate empirically the various sums of squares in the analysis of variance table. Searle(1977) and Searle et.al.(1981) developed a method of computing R($\alpha
\mu, \beta$) and R($\beta
\mu, \alpha$) by the use of partitioned matrix of X'X for the model of no interaction, interchanging the columns of X in order of
and accordingly the elements in b. An alternative way of computing R($\alpha
\mu, \beta$), R($\beta
\mu, \alpha$) and R($\gamma
\mu, \alpha, \beta$) without interchanging the columns of X has been found by means of,
. It is true that $R(\alpha
\mu,\beta,\gamma)\Sigma = SSA_W and R(\beta
\mu,\alpha,\gamma)\Sigma = SSB_W$ where
and means analysis and $R(\gamma
\mu,\alpha,\beta) = R(\gamma
\mu,\alpha,\beta)\Sigma$ for the data without empty cell, but not for the data with empty cell(s). It is also noticed that for the datd with empty cells under W - restrictions $R(\alpha
\mu,\beta,\gamma)_W = R(\mu,\alpha,\beta,\gamma)_W - R(\mu,\alpha,\beta,\gamma)_W = R(\alpha
\mu) and R(\beta
\mu,\alpha,\gamma)_W = R(\mu,\alpha,\beta,\gamma)_W - R(\mu,\alpha,\beta,\gamma)_W = R(\beta
\mu) but R(\gamma
\mu,\alpha,\beta)_W = R(\mu,\alpha,\beta,\gamma)_W - R(\mu,\alpha,\beta,\gamma)_W \neq R(\gamma
\mu,\alpha,\beta)$. The hypotheses
commonly tested were examined in the relation with the corresponding sums of squares for $R(\alpha
\mu,\alpha,\gamma), and R(\gamma
\mu,\alpha,\beta)$ under the restrictions.
Comparison of parameter estimation methods for time series models in the presence of outliers
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 255~268
We propose an iterated interpolation approach for the estimation fo time series parameters in the presence of outliers. The proposed approach iterates the parameter estimation stage and the outlier detection stage until no further outliers are detected. For the detection of outliers, interpolation diagnostic is applied, where the atypical observations by the one-step-ahead predictor instead of downweighting is also proposed. The performance of the proposed estimation methods is compared with other robust estimation methods by simulation study. It is observed that the iterated interpolation approach performs reasonably well is general, especially for single AO case and large
in absolute values.
A statistical consideration on the number of occurrences of langerhans cells
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 271~282
A statistical method to investigate the relationship between the occurrence of Langerahans cells and neoplastic transformation of uterine cerivx. The best fitting submodel which satisfies the selection criterion similar in type to AIC is selected among the possible submodels based on Poisson probability models. A bootstrap method is used to approximate the sampling distribution of the selection criterion and the usual normal approximation is used to find the asymptotic distribution of the estimated rates.
Updating algorithms in statistical computations
Korean Journal of Applied Statistics, volume 5, issue 2, 1992, Pages 283~292
Updating algorithms are studied for the basic statistics (mean, variance). For a linear model, a recursive formulae for least squares estimators of regression coefficients, residual sum of squares and variance-covariance matrix are also studied. Hotelling's $T^2$ statistics can be calculated recursively using the recursive formulae of mean vector and variance-covariance matrix without computing the sample variance-covariance matrix at each stage.