• Journal of the Korean Statistical Society
• /
• v.25 no.1
• /
• pp.67-80
• /
• 1996

Factor analysis is a multivariate technique for describing the in-terrelationship among many variables in terms of a few underlying but unobservable random variables called factors. There are various approaches for this factor analysis. In particular, principal factor analysis is one of the most popular methods. This follows the mathematical algorithm of the principal component analysis based on the singular value decomposition. But it is known that the singular value decomposition is not resistant, i.e., it is very sensitive to small changes in the input data. In this article, using the resistant singular value decomposition of Choi and Huh (1994), we derive a resistant principal factor analysis relatively little influenced by notable observations.

• The Korean Journal of Applied Statistics
• /
• v.26 no.6
• /
• pp.933-942
• /
• 2013

Common factor analysis and principal component analysis represent two technically distinctive approaches to exploratory factor analysis. Much of the psychometric literature recommends the use of common factor analysis instead of principal component analysis. Nonetheless, factor analysts use principal component analysis more frequently because they believe that principal component analysis could yield (relatively) less accurate estimates of factor loadings compared to common factor analysis but most often produce similar pattern of factor loadings, leading to essentially the same factor interpretations. A simulation study is conducted to evaluate the relative performance of these two approaches in terms of factor pattern recovery under different experimental conditions of sample size, overdetermination, and communality.The results show that principal component analysis performs better in factor recovery with small sample sizes (below 200). It was further shown that this tendency is more prominent when there are a small number of variables per factor. The present results are of practical use for factor analysts in the field of marketing and the social sciences.

• Journal of Korean Society of Rural Planning
• /
• v.12 no.1 s.30
• /
• pp.37-48
• /
• 2006

The objectives of this study were to develop an evaluation method of regional vulnerability to agricultural drought and to classify the vulnerability patterns. In order to test the method, 24 city or county areas of Gyeonggi-do were chose. First, statistic data and digital maps referred for agricultural drought were defined, and the input data of 31 items were set up from 5 categories: land use factor, water resource factor, climate factor, topographic and soil factor, and agricultural production foundation factor. Second, for simplification of the factors, principal component analysis was carried out, and eventually 4 principal components which explain about 80.8% of total variance were extracted. Each of the principal components was explained into the vulnerability components of scale factor, geographical factor, weather factor and agricultural production foundation factor. Next, DVIP (Drought Vulnerability Index for Paddy), was calculated using factor scores from principal components. Last, by means of statistical cluster analysis on the DVIP, the study area was classified as 5 patterns from A to E. The cluster A corresponds to the area where the agricultural industry is insignificant and the agricultural foundation is little equipped, and the cluster B includes typical agricultural areas where the cultivation areas are large but irrigation facilities are still insufficient. As for the cluster C, the corresponding areas are vulnerable to the climate change, and the D cluster applies to the area with extensive forests and high elevation farmlands. The last cluster I indicates the areas where the farmlands are small but most of them are irrigated as much.

• Journal of Environmental Science International
• /
• v.7 no.2
• /
• pp.171-176
• /
• 1998

This study was conducted to evaluate water quality utilizing principal component analysis in the Nakdong River Estuary. From the results of analysis, water quality in the Nakdong River Estuary could be explained up to 65.3 Percente by three factors which were Included In river loadlnwastes from the Nakdong River and rainfalls : 39.1%1, sediment resuspension(13.7BS) and metabolism(12.5%). In the eastern part of estuary In flowing the Nakdong River, river loading factor score(factor 1 Pas higher than that In western part. Sediment resuspension factor score(factor 2) was high in shallow water, while metabolism factor score(factor 3) was high in deeper water. For seasonal variations of factors score, factor 1 was h19h- 1y related to rainfall season.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.4
• /
• pp.773-787
• /
• 2003

Principal component analysis, factor analysis and cluster analysis have been performed to analyze the relationship between air pollutants and meteorological variables measured in 1999 in Seoul. In principal analysis, the first principal has been shown the contrast effect between $O_3$ and the other pollutants, the second principal has been shown the contrast effect between CO, $SO_2$, $NO_2$ and $O_3$, PM10, TSP. In factor analysis, the first factor has been found as PM10, TSP, $NO_2$ concentrations which are related with suspended particulates. As a result of cluster analysis, three clusters respectively have represented different air pollution levels, seasonal characteristics of air pollutants and meteorological situations.

• Communications for Statistical Applications and Methods
• /
• v.6 no.1
• /
• pp.33-42
• /
• 1999

A method of detecting influential observations in principal factor analysis is suggested. it is based on a perturbation of the empirical distribution function and an adoption of the local influence method. An illustrative example is given.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.3
• /
• pp.293-298
• /
• 2015

The vessel encounter data collected from the vessel trajectories in the maritime traffic situation is possible to analyze vessel collision and near-collision risk using statistical method. In this study, analyzing variables extracted from the vessel encounter data using factor analysis, we determine main factors effecting vessel collision risk from vessel encounter data. In order to calculate each factor, it used principal component analysis for factor analysis after normalization and standardization of vessel encounter variables. As a result of the factor analysis, main effect factors are summarized into the vessel approach factor and collision avoidance variance factor.

• Journal of Labour Economics
• /
• v.28 no.1
• /
• pp.61-82
• /
• 2005

We investigate wage differentials in Korea in the manufacturing industry, as well as factors affecting structural change in wage determination for the pre- and post-financial crisis regimes. We use the 1995 and 1999 data from the Survey Report on the Wage Structure (SRWS) from the Ministry of Labor. Principal components regression analysis is used to tackle multicollinearity. We employ factor analysis to reduce a set of variables to a smaller number, which contain observed and latent variables. Our empirical investigation provide evidences for changes in wages structure between 1995 and 1999. In 1995, the job quality factor is the most critical in the determination of wages, while in 1999, the industry attributes factor impacts greatly on the wages.

• Journal of applied mathematics & informatics
• /
• v.14 no.1_2
• /
• pp.377-386
• /
• 2004

A technique that provides prediction intervals based on a model called an empirical linear model is discussed. The technique, high-dimensional empirical linear prediction (HELP), involves principal component analysis, factor analysis and model selection. HELP can be viewed as a technique that provides prediction (and confidence) intervals based on a factor analysis models do not typically have justifiable theory due to nonidentifiability, we show that the intervals are justifiable asymptotically.

• Journal of Korea Multimedia Society
• /
• v.10 no.8
• /
• pp.981-996
• /
• 2007

In this paper, we describe a method that can recognize gestures by obtaining motion features information with principal factor analysis from sequential gesture images. In the algorithm, firstly, a two dimensional silhouette region including human gesture is segmented and then geometric features are extracted from it. Here, global features information which is selected as some meaningful key feature effectively expressing gestures with principal factor analysis is used. Obtained motion history information representing time variation of gestures from extracted feature construct one gesture subspace. Finally, projected model feature value into the gesture space is transformed as specific state symbols by grouping algorithm to be use as input symbols of HMM and input gesture is recognized as one of the model gesture with high probability. Proposed method has achieved higher recognition rate than others using only shape information of human body as in an appearance-based method or extracting features intuitively from complicated gestures, because this algorithm constructs gesture models with feature factors that have high contribution rate using principal factor analysis.