• Journal of the Korean Statistical Society
• /
• v.33 no.3
• /
• pp.271-282
• /
• 2004

We discuss complete convergence of weighted sums for arrays of ø-mixing random variables. As application, we obtain the complete convergence of moving average processes for ø-mixing random variables. The result of Baum and Katz (1965) as well as the result of Li et al. (1992) on iid case are extended to ø-mixing setting.

• Journal of Korean Society for Quality Management
• /
• v.19 no.2
• /
• pp.172-180
• /
• 1991

The null hypothesis being tested by $the{\bar{X}}$ control chart is that the process is in control at a quality level ${\mu}o$. An ${\bar{X}}control$ chart is a tool for detecting process average changes due to assingnable causes. The major weakness of $the{\bar{X}}$ control chart is that it is relatively insensitive to small changes in the population mean. This paper presents one way to remedy this weakness is to allow each plotted value to depend not only on the most recent subgroup average but on some of the other subgroup averages as well. Two approaches for doing this are based on (1) moving averages and (2) exponentially weighted moving averages of forecasting method.

• Phonetics and Speech Sciences
• /
• v.2 no.4
• /
• pp.145-151
• /
• 2010

In this paper, a robust feature compensation method is proposed for improving the performance of speech recognition. The proposed method is incorporated into the auto-regressive moving average (ARMA) based feature compensation. We employ variable weights for the ARMA filter according to the degree of speech activity, and pass the normalized cepstral sequence through the weighted ARMA filter. Additionally when normalizing the cepstral sequences in training, the cepstral means and variances are estimated from total training utterances. Experimental results show the proposed method significantly improves the speech recognition performance in the noisy and reverberant environments.

• Communications for Statistical Applications and Methods
• /
• v.19 no.3
• /
• pp.367-379
• /
• 2012

In baseball rankings, prediction has been a subject of interest for baseball fans. To predict these rankings, (based on 2011 data from Korea Professional Baseball records) the arithmetic mean method, the weighted average method, principal component analysis, and principal component regression analysis is presented. By standardizing the arithmetic average, the correlation coefficient using the weighted average method, using principal components analysis to predict rankings, the final model was selected as a principal component regression model. By practicing regression analysis with a reduced variable by principal component analysis, we propose a rank predictability model of a pitcher part, a batter part and a pitcher batter part. We can estimate a 2011 rank of pro-baseball by a predicted regression model. By principal component regression analysis, the pitcher part, the other part, the pitcher and the batter part of the ranking prediction model is proposed. The regression model predicts the rankings for 2012.

• Industrial Engineering and Management Systems
• /
• v.7 no.2
• /
• pp.101-112
• /
• 2008

As charts to monitor the process fraction defectives, P, in the high-yield processes, Mishima et al. (2002) discussed a synthetic chart, the Synthetic CS chart, which integrates the CS (Confirmation Sample)$_{CCC(\text{Cumulative Count of Conforming})-r}$ chart and the CCC-r chart. The Synthetic CS chart is designed to monitor quality characteristics in real-time. Recently, Kotani et al. (2005) presented the EWMA (Exponentially Weighted Moving-Average)$_{CCC-r}$ chart, which considers combining the quality characteristics monitored in the past with one monitored in real-time. In this paper, we present an alternative chart that is more superior to the $EWMA_{CCC-r}$ chart. It is an integration of the $EWMA_{CCC-r}$ chart and the CCC-r chart. In using the proposed chart, the quality characteristic is initially judged as either the in-control state or the out-of-control state, using the lower and upper control limits of the $EWMA_{CCC-r}$ chart. If the process is not judged as the in-control state by the $EWMA_{CCC-r}$ chart, the process is successively judged, using the $EWMA_{CCC-r}$ chart. We compare the ANOS (Average Number of Observations to Signal) of the proposed chart with those of the $EWMA_{CCC-r}$ chart and the Synthetic CS chart. From the numerical experiments, with the small size of inspection items, the proposed chart is the most sensitive to detect especially the small shifts in P among other charts.

• The Korean Journal of Applied Statistics
• /
• v.37 no.2
• /
• pp.177-189
• /
• 2024

The run length is defined as the number of samples or subgroups taken before the control chart statistic exceeds the control limits. Because the distribution of run length is typically asymmetric and has a large variability, it may not be appropriate to use ARL (average run length) alone to design control charts and evaluate performance. In this paper, we introduce the concept of percentile (PL)-based design of control charts, and propose the procedure for PL-based design of EWMA (exponentially weighted moving average) charts. For the PL-based design of EWMA, we present a fitted function for the control chart coefficient, given specific percentile parameters. Additionally, we perform simulations to compare the proposed design with the ARL-based design. The simulation results show that the proposed design yields improvements in monitoring in-control processes while maintaining the ability to detect out-of-control performance.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.4
• /
• pp.1877-1885
• /
• 2015

This paper presents a new Nearest-Neighbors based weighted representation for images and weighted K-Nearest-Neighbors (WKNN) classifier to improve the precision of image classification using the Bag of Visual Words (BOVW) based models. Scale-invariant feature transform (SIFT) features are firstly extracted from images. Then, the K-means++ algorithm is adopted in place of the conventional K-means algorithm to generate a more effective visual dictionary. Furthermore, the histogram of visual words becomes more expressive by utilizing the proposed weighted vector quantization (WVQ). Finally, WKNN classifier is applied to enhance the properties of the classification task between images in which similar levels of background noise are present. Average precision and absolute change degree are calculated to assess the classification performance and the stability of K-means++ algorithm, respectively. Experimental results on three diverse datasets: Caltech-101, Caltech-256 and PASCAL VOC 2011 show that the proposed WVQ method and WKNN method further improve the performance of classification.

• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.451-457
• /
• 2002

We derive the strong laws of large numbers for weighted averages of partial sums of random variables which are either associated or negatively associated. Our theorems extend and generalize strong law of large numbers for weighted sums of associated and negatively associated random variables of Matula(1996; Probab. Math. Statist. 16) and some results in Birkel(1989; Statist. Probab. Lett. 7) and Matula (1992; Statist. Probab. Lett. 15 ).

• The Korean Journal of Applied Statistics
• /
• v.28 no.5
• /
• pp.1003-1011
• /
• 2015

We studied indices to explain runs lost for Korean professional baseball teams. Kim and Kim (2014) studied batting indices to explain run productivity of teams; subsequently, we studied fielding indices to explain runs lost. We considered several combined indices made by combining fielding indices closely connected with the runs lost of teams. Data analysis from all games in the regular seasons of 1982~2014 show that weighted WPH (defined as weighted average of WHIP and number of home runs allowed per game) best explain runs lost. Weighted WPH consisting of WHIP (with weight 81%) and number of home runs allowed per game (with weight 19%) was found optimal weighted WPH having correlation coefficient 0.95033 with average runs lost per game. Analysis by chronological periods gave results not much different.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.4
• /
• pp.563-568
• /
• 2007

This paper presents a methodology for predicting nonlinear time series based on the neural network with weighted fuzzy membership functions (NEWFM). The degree of classification intensity is obtained by bounded sum of weighted fuzzy membership functions extracted by NEWFM, then weighted average defuzzification is used for predicting nonlinear time series. The experimental results demonstrate that NEWFM has the classification capability of 92.22% against the target class of GDP. The time series created by NEWFM model has a relatively close approximation to the GDP which is a typical business cycle indicator, and has been proved to be a useful indicator which has the turning point forecasting capability of average 12 months in the peak point and average 6 months in the trough point during 5th to 8th cyclical period. In addition, NEWFM measures the efficiency of the economic indexes by the feature selection and enables the users to forecast with reduced numbers of 7 among 10 leading indexes while improving the classification rate from 90% to 92.22%.