Factor Extraction Method Research Articles

Artificial neural network based peak load forecasting using Levenberg–Marquardt and quasi-Newton methods

Daily electrical peak-load forecasting has been done using the feedforward neural network based on the Levenberg-Marquardt back-propagation algorithm, Broyden-Fletcher-Goldfarb-Shanno back-propagation algorithm and one-step secant backpropagation algorithm by incorporating the effect of eleven weather parameters, the type of day and the previous day peak load information. To avoid the trapping of the network into a state of local minima, the optimisation of user-defined parameters viz. learning rate and error goal has been performed. Training data set has been selected using a growing window concept and is reduced as per the nature of the day and the season for which the forecast is made. For redundancy removal in the input variables, reduction of the number of input variables has been done by the principal component analysis method of factor extraction. The resultant data set is used for the training of a three-layered neural network. To increase the learning speed, the weights and biases are initialised according to the Nguyen and Widrow method. To avoid over-fitting, an early stopping of training is done at the minimum validation error.

Extracting factors from heteroskedastic asset returns

This paper proposes an alternative to the asymptotic principal components procedure of Connor and Korajczyk (J. Financial Econom. 15 (1986) 373) that is robust to time series heteroskedasticity in the factor model residuals. The new method is simple to use and requires no assumptions stronger than those made by Connor and Korajczyk. It is demonstrated through simulations and analysis of actual stock market data that allowing heteroskedasticity sometimes improves the quality of the extracted factors quite dramatically. Over the period from 1989 to 1993, for example, a single factor extracted using the Connor and Korajczyk method explains only 8.2% of the variation of the CRSP value-weighted index, while the factor extracted allowing heteroskedasticity explains 57.3%. Accounting for heteroskedasticity is also important for tests of the APT, with p-values sometimes depending strongly on the factor extraction method used.

Unification of stress intensity factor (SIF) extraction methods with an h‐adaptive finite element scheme

AbstractAn h‐adaptive finite element procedure consisting of recently developed mesh generator (advancing front type and exclusively suitable for two‐dimensional domains containing cracks), ZZ‐error estimator and a refinement strategy based on strain energy concentration index have been used to refine the meshes of domains containing cracks adaptively to a target error of 3 per cent. Seven mixed‐mode SIF extraction methods have been coupled with the present h‐adaptive finite element analysis in order to examine the performance of each method in an environment of randomly changing meshes. Based on the performance of each method the SIF values of a mixed‐mode crack problem that does not have analytical solution have been suggested. Copyright © 2001 John Wiley & Sons, Ltd.

The Equivalence of SF-36 Summary Health Scores Estimated Using Standard and Country-Specific Algorithms in 10 Countries: Results from the IQOLA Project

Data from general population surveys ( n = 1771 to 9151) in nine European countries (Denmark, France, Germany, Italy, the Netherlands, Norway, Spain, Sweden, and the United Kingdom) were analyzed to test the algorithms used to score physical and mental component summary measures (PCS-36/MCS-36) based on the SF-36 Health Survey. Scoring coefficients for principal components were estimated independently in each country using identical methods of factor extraction and orthogonal rotation. PCS-36 and MCS-36 scores were also estimated using standard (U.S.-derived) scoring algorithms, and results were compared. Product-moment correlations between scores estimated from standard and country-specific scoring coefficients were very high (0.98 to 1.00) for both physical and mental health components in all countries. As hypothesized for orthogonal components, correlations between physical and mental components within each country were very low (0.00 to 0.12) for both estimation methods. Mean scores for PCS-36 differed by as much as 3.0 points across countries using standard scoring, and mean scores for MCS-36 differed across countries by as much as 6.4 points. In view of the high degree of equivalence observed within each country, using standard and country-specific algorithms, we recommend use of standard scoring algorithms for purposes of multinational studies involving these 10 countries.

A comparison of factor analytic methods of obtaining cardiovascular autonomic components for the assessment of mental workload

The diagnosticity of heart rate for mental workload assessment can be improved with an autonomic space model of sympathetic and parasympathetic nervous system influences on the heart. Methods of deriving autonomic components to estimate the underlying sympathetic and parasympathetic activity needed to identify autonomic modes of control for heart period were examined. Alternative factor extraction and rotation methods were compared using data from a laboratory study that measured the electrocardiogram and impedance cardiogram simultaneously. Principal components analysis with varimax rotation was found to validly estimate the sympathetic and parasympathetic activity when computed on multiple psychophysiological measures obtained from raw EKG data. Use of the method was illustrated with single and dual mental arithmetic and secondorder compensatory manual tracking tasks. Different autonomic modes of control were found for divided attention that were not evident in heart period. Uncoupled sympathetic activation was indicated for divided attention when the mental arithmetic tasks were added to the tracking single tasks, but uncoupled parasympathetic inhibition was indicated for divided attention when the tracking tasks were added to the mental arithmetic single tasks.

A modeling-based factor extraction method for determining spatial heterogeneity of Ga-68 EDTA kinetics in brain tumors

The ROI method applied to Ga-68 EDTA PET dynamic study data for the quantitative determination of brain tumor BBB permeability assumes that the tumor is homogeneous in terms of Ga-68 EDTA kinetics, even though it is known that brain tumors are highly heterogeneous in structure. The relatively high image noise of Ga-68 PET studies have prevented the examination of Ga-68 EDTA kinetics by nonlinear regression on a single pixel basis. In this study, we have developed an efficient and effective method to separate brain tumor tissue into sub-regions with different Ga-68 EDTA kinetics on a pixel-by-pixel basis. Computer simulation and ten Ga-68 EDTA PET patient studies were used to evaluate the performance of the new method. During a PET dynamic study (total 64 min), 20-25 arterial samples were taken for the input function. Whole-tumor ROIs were defined on T1-weighted MRI images and then copied to the registered PET dynamic images to measure whole-tumor time activity curves. The method uses a two-compartment model to extract three component factors (vascular component, fast and slow component factors) from whole-tumor kinetics by model fitting. The kinetics in each pixel were expressed as a linear combination of the three factors. The three coefficients in the expression can be estimated by the linear least-square method and produce three factor images corresponding, respectively, to the permeability of the fast and the slow component factors and the plasma volume. Whole-tumor regions were separated into two regions-one with mainly fast kinetics that was evident in the fast factor images and one with slow kinetics that was evident in slow factor images. The two regions have markedly different uptake (0.036/spl plusmn/0.015 ml/min/g and 0.009/spl plusmn/0.006 ml/min/g for fast and slow kinetic sub-regions, respectively) and clearance rates (0.22/spl plusmn/0.15 /min and 0.023/spl plusmn/0.021 /min for fast and slow sub-regions, respectively). The overlap of the two resulting sub-regions is small (3.5/spl plusmn/1.7% of the two regions). Computer simulation and patient studies show that the method is robust for a wide range of noise levels. This method has combined the advantages of statistical factor analysis and the modeling approach.

Divergent Validity and the Measurement of Processing Capacity

I do not disagree with most of the substantive conclusions that Steven Pulos draws, based on his reanalyses of Case and Globerson (1974) and Morra’ s (1994) data. My problem is with his epistemological presuppositions. He writes as if the classic epistemology of exploratory factor analysis is somewhat truer and more rigorous than the epistemology of developmental constructivism. In classical exploratory factor analysis a main semantic assumption is that the obtained factors are indexing causal determinants that can explain the matrix of correlations. However, if we look at what these factors actually are in terms of the operations involved in their extraction, factors appear instead as higher-order descriptive response variables: They represent components of covariation patterns among tests, in terms of subjects’ own responses— relational “ empirical invariants” or latent variables. Thus, even more clearly than correlations themselves, e rst-order factors represent quantie ed relations of equivalence among tests, in terms of the subjects’ responses. One might concede that factors must then stand for relations of coexistence in the tests loading on them— relations among semantic/qualitative facets or aspects that coexist in tests loading on the same factor; these are facets that the tests share. But nothing in the technique of factor analysis tells us which these facets actually are. To answer these questions, factor analysts have traditionally used phenomenology, often a home-grown variety, perhaps a nai ve one. Factor analysts have not until recently (e.g. Frederiksen, Mislevy, & Bejar, 1993) been interested in subjective/rational task analysis— a form of applied phenomenology which, in combination with suitable developmental-process and individual-difference elements, lends itself to the semantic clarie cation of empirical tests. The hierarchical structure of factors proposed by classical factor analysts is predicated on the method of factor extraction, and implies

Denervation, but not decentralization, reduces nerve growth factor content of the mesenteric artery.

In the present study we applied an improved nerve growth factor (NGF) extraction method to examine the effects of denervation and sympathetic decentralization on NGF levels in vascular tissue. Adult male Wistar Kyoto rats underwent mesenteric arterial denervation or splanchnic nerve transection. Four days after operation, animals were killed, and the mesenteric artery and coeliac-superior mesenteric ganglia were removed. The arterial adventitia was stripped from the media to measure NGF levels in nerve and smooth muscle separately. A high concentration of NGF was detected in the normal artery, 90% of which was in the adventitial layer. Surgical denervation significantly reduced the NGF levels in the artery and ganglia by 78 and 71%, respectively. However, within the artery the level of NGF was reduced in the adventitia but not in the media. Thus, the large reduction of NGF content resulted from the loss of nerve plexus from the artery. In contrast, decentralization did not alter the NGF content in the artery, in either the adventitia or media. Our results are in marked contrast to previous studies reporting elevated levels of NGF following denervation. This discrepancy is explained by the ability of our new procedure to extract much greater amounts of NGF from the tissue.

Factor structure of the WISC‐R for Mexican‐American students referred for intellectually gifted assessment

This study assessed the factor structure of the WISC‐R with Mexican‐American students referred for intellectually gifted assessment. The students included 68 Mexican‐American students who were referred for evaluation for an intellectually gifted program. The 12 WISC‐R subtest scaled scores were used as the variables to be factored. The method of factor extraction used was the maximum likelihood analysis followed by a varimax rotation. On each of the previously‐identified three factors, differences were observed.

A First Course in Factor Analysis

Contents: Preface to the First Edition. Preface to the Second Edition. Introduction. The Factor Analytic Model. Factor Extraction by the Centroid Method. Methods of Factor Extraction. Orthogonal Hand Rotations. Oblique Hand Rotations. Simple Structure and Other Rotational Criteria. Planning the Standard Design Factor Analysis. Alternate Designs in Factor Analysis. Interpretation and Application of Factor Analytic Results. Development of the Comrey Personality Scales: An Example of the Use of Factor Analysis. Confirmatory Factor Analysis. Structural Equation Models. Computer Programs.

Hyball: A Method for Subspace-Constrained Factor Rotation.

Want to rotate the latent variables you have found by some method of factor extraction to oblique simple structure while leaving invariant selected axes or subspaces in your initial solution? Here's how.

Dimensions of child-rearing practices. Factor structure of the EMBU.

Measurement of the dimensions of parental-rearing practices was undertaken using an English form of the EMBU inventory, which was administered to 282 non-clinical subjects. Results indicated that eight of the 14 subscales of the EMBU could be identified on oblique factor analysis, and that six other dimensions were specific to Mother or Father. Results suggest that method of factor extraction plays an important part in number of dimensions of child-rearing practices identified, and that it may be appropriate to assume that dimensions of parental rearing are correlated rather than independent.

The language of low back pain: Factor structure of the McGill pain questionnaire

Low back pain patients' responses to the McGill Pain Questionnaire were factor analyzed using a method of factor extraction which, relative to the procedures employed in previous investigations, minimized distortion of the factor solution. Four factors were found which accounted for 51% of the total variance. Three of the factors were composed solely of sensory, affective and evaluate descriptor subclasses, respectively. The fourth was defined by both sensory and affective subclasses. The results provided strong support for the continued use of the sensory, affective and evaluate pain rating indices. Suggestions regarding replication of the present study and use of pain language factor scales for predictive purposes were discussed.

Evaluation of factor analytic research procedures by means of simulated correlation matrices

In order to study the effectiveness of factor analytic methods, a procedure was developed for computing simulated correlation matrices which are more similar to real data correlation matrices than are those matrices computed from the factor analysis structural model. In the present investigation, three methods of factor extraction were studied as applied to 54 simulated correlation matrices which varied in proportion of variance derived from a major factor domain, number of factors in the major domain, and closeness of the simulation procedure to the factor analysis structural model. While the factor extraction methods differed little from one another in quality of results for matrices more dissimilar to the factor analytic model, major differences in quality of results were associated with fewer factors in the major domain, higher proportion of variance from the major domain, and closeness of the simulation procedure to the factor analysis structural model.