Inference in regression models of heavily skewed alcohol use data: A comparison of ordinary least squares, generalized linear models, and bootstrap resampling.

Analysis of alcohol use data and other low base rate risk behaviors using ordinary least squares regression models can be problematic. This article presents 2 alternative statistical approaches, generalized linear models and bootstrapping, that may be more appropriate for such data. First, the basic theory behind the approaches is presented. Then, using a data set of alcohol use behaviors and consequences, results based on these approaches are contrasted with the results from ordinary least squares regression. The less traditional approaches consistently demonstrated better fit with model assumptions, as demonstrated by graphical analysis of residuals, and identified more significant variables potentially resulting in theoretically different interpretations of the models of alcohol use. In conclusion, these models show significant promise for furthering the understanding of alcohol-related behaviors. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Fitting intracranial self-stimulation data with growth models.

Until now, the problem of fitting self-stimulation rate-frequency functions has been dealt with by using linear models applied to the linear portion of the empirical curve. In this article, an alternative procedure is presented, together with three sigmoid growth models that seem to accurately fit rate-frequency data. From any of these models, it is possible to compute the two indices of stimulation efficacy in use in the parametric study of brain stimulation reward (M?? and θ?), in addition to the inflection point of the curve, which can be used as an alternative to M??. Important relations allowing initial estimation of each parameter are provided, allowing use of computer programs derived from the Gauss-Newton algorithm for nonlinear regression. The considerations relevant to the choice of a nonlinear model are discussed in terms of each efficacy index. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Sequential effects in magnitude scaling: Models and theory.

Research on sequential effects in magnitude scaling is reviewed, and its implications about the adequacy of current time series regression models is discussed. A regression model that unifies what at first appears to be contradictory results is proposed. Theoretical models of judgment and perception are introduced, and their relation to alternative regression models is clarified. A theoretical model of relative judgment that clarifies the role of judgmental error and frames of reference in magnitude scaling is examined in detail. Four experiments that test the model are presented. The results, along with recent results presented by L. M. Ward (see record 1987-24003-001) provide support for the model. The importance of being explicit about the relation of theoretical models to regression models and about the role of error in these models is discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Soil Salinity Sampling Strategy Using Spatial Modeling Techniques, Remote Sensing, and Field Data

The aim of this study was to develop a methodology for collecting soil salinity samples. The objectives of this paper are to: (1) estimate the number of soil salinity samples needed to capture the variability in the soil salinity data with high accuracy; and (2) compare two types of satellite images with different resolutions: Ikonos with 4?m resolution and Landsat 5 with 30?m resolution. To achieve these objectives, two satellite images were acquired (one for Ikonos and one for Landsat 5) to evaluate the correlation between the measured soil salinity and remote sensing data. From the observed data, three subsets were randomly extracted with each subset containing: 75, 50, and 25% of the data for each field. These three subsets were then used in the modeling process. Ordinary least squares (OLS) (i.e., multiple regression) was used to explore the coarse-scale variability in soil salinity as a function of the Ikonos and Landsat 5 bands. A stepwise procedure was used to identify the best subset of satellite bands to include in the regression models that minimized the Akakie information criteria (AIC). Then, the spatial structure of the residuals from the OLS models were described using sample variogram models. The variogram model with the smallest AIC was selected to describe the spatial dependencies in the soil salinity data. If the residuals were spatially correlated, ordinary kriging was used to model the spatial distribution of soil salinity in the fields. A tenfold cross validation was used to estimate the prediction error for soil salinity. To evaluate the effectiveness of the models, various measures of the prediction error were computed. This study provides an accurate methodology that can be used by researchers in reducing the number of soil samples that need to be collected. This is especially valuable in projects that last several years. The results of this study suggest that the number of soil samples that need to be collected and therefore their cost can be significantly reduced and soil salinity estimation can be significantly improved by using kriging. The results also show that the Ikonos image performed better than the Landsat 5 image.     

Calcium and mechanically induced potentials in fibroblasts of rat atrium

Pattern-mixture models stratify incomplete data by the pattern of missing values and formulate distinct models within each stratum. Pattern-mixture models are developed for analyzing a random sample on continuous variables y(1), y(2) when values of y(2) are nonrandomly missing. Methods for scalar y(1) and y(2) are here generalized to vector y(1) and y(2) with additional fixed covariates x. Parameters in these models are identified by alternative assumptions about the missing-data mechanism. Models may be underidentified (in which case additional assumptions are needed), just-identified, or overidentified. Maximum likelihood and Bayesian methods are developed for the latter two situations, using the EM and SEM algorithms, direct and interactive simulation methods. The methods are illustrated on a data set involving alternative dosage regimens for the treatment of schizophrenia using haloperidol and on a regression example. Sensitivity to alternative assumptions about the missing-data mechanism is assessed, and the new methods are compared with complete-case analysis and maximum likelihood for a probit selection model.     

Power and sample size calculations for studies involving linear regression

This article presents methods for sample size and power calculations for studies involving linear regression. These approaches are applicable to clinical trials designed to detect a regression slope of a given magnitude or to studies that test whether the slopes or intercepts of two independent regression lines differ by a given amount. The investigator may either specify the values of the independent (x) variable(s) of the regression line(s) or determine them observationally when the study is performed. In the latter case, the investigator must estimate the standard deviation(s) of the independent variable(s). This study gives examples using this method for both experimental and observational study designs. Cohen's method of power calculations for multiple linear regression models is also discussed and contrasted with the methods of this study. We have posted a computer program to perform these and other sample size calculations on the Internet (see http://www.mc.vanderbilt.edu/prevmed/psintro+ ++.htm). This program can determine the sample size needed to detect a specified alternative hypothesis with the required power, the power with which a specific alternative hypothesis can be detected with a given sample size, or the specific alternative hypotheses that can be detected with a given power and sample size. Context-specific help messages available on request make the use of this software largely self-explanatory.     

Distinguishing between moderator and quadratic effects in multiple regression.

Moderated regression analysis is commonly used to test for multiplicative influences of independent variables in regression models. D. Lubinski and L. G. Humphreys (1990) have shown that significant moderator effects can exist even when stronger quadratic effects are present. They recommend comparing effect sizes associated with both effect types and selecting the model that yields the strongest effect. The authors show that this procedure of comparing effect sizes is biased in favor of the moderated model when multicollinearity is high because of the differential reliability of the quadratic and multiplicative terms in the regression models. Fortunately, levels of multicollinearity under which this bias is most problematic may be outside the range encountered in many empirical studies. The authors discuss causes and implications of this phenomenon as well as alternative procedures for evaluating structural relationships among variables. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Artificial Intelligence-Based Inductive Models for Prediction and Classification of Fecal Coliform in Surface Waters

This paper describes the use of inductive models developed using two artificial intelligence (AI)-based techniques for fecal coliform prediction and classification in surface waters. The two AI techniques used include artificial neural networks (ANNs) and a fixed functional set genetic algorithm (FFSGA) approach for function approximation. While ANNs have previously been used successfully for modeling water quality constituents, FFSGA is a relatively new technique of inductive model development. This paper will evaluate the efficacy of this technique for modeling indicator organism concentrations. In scenarios where process-based models cannot be developed and/or are not feasible, efficient and effective inductive models may be more suitable to provide quick and reasonably accurate predictions of indicator organism concentrations and associated water quality violations. The relative performance of AI-based inductive models is compared with conventional regression models. When raw data are used in the development of the inductive models described in this paper, the AI models slightly outperform the traditional regression models. However, when log transformed data are used, all inductive models show comparable performance. While the work validates the strength of simple regression models, it also validated FFSGA to be an effective technique that competes well with other state-of-the-art and complex techniques such as ANNs. FFSGA comes with the added advantage of resulting in a simple, easy to use, and compact functional form of the model sought. This work adds to the limited amount of research on the use of data-driven modeling methods for indicator organisms.     

Developing Regression Models for Predicting Pan Evaporation from Climatic Data—A Comparison of Multiple Least-Squares, Principal Components, and Partial Least-Squares Approaches

Regression models for predicting daily pan evaporation depths from climatic data were developed using three multivariate approaches: multiple least-squares regression (MLR), principal components regression (PCR), and partial least-squares (PLS) regression. The objective was to compare the prediction accuracies of regression models developed by these three approaches using historical climatic datasets of four Indian sites that are located in distinctly different climatic regimes. In all cases (three approaches applied to four climatic datasets), regression models were developed using a part of the data and subsequently validated with the remaining data. Results indicated that although performances of the regression models varied from one climate to another, more or less similar prediction accuracies were obtained by all three approaches, and it was difficult to identify the best approach based on performance statistics. However, the final forms of the regression models developed by the three approaches differed substantially from one another. In all cases, the models derived using PLS contained the smallest number of predictor variables; between two to three out of a possible maximum of six predictor variables. The MLR approach yielded models with three to six predictor variables, and PCR models included all six predictor variables. This implies that the PLS regression models are the most parsimonious in terms of input data required for estimating epan from climate variables, and yet yield predictions that are almost as accurate as the more data-intensive MLR and PCR models.     

Multiple monotone regression.

Multiple monotone regression is a set of models and methods that can be applied when researchers do not think that the relationships among their measured variables can be adequately described by a multiple linear regression model. The relations among weak monotonic, additive, and additive monotonic models are discussed. As an example, the models are applied to data from the present author's (see record 1982-27256-001) study of visual information processing. (17 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Longitudinal tests of competing factor structures for the Rosenberg Self-Esteem Scale: Traits, ephemeral artifacts, and stable response styles.

Self-esteem, typically measured by the Rosenberg Self-Esteem Scale (RSE), is one of the most widely studied constructs in psychology. Nevertheless, there is broad agreement that a simple unidimensional factor model, consistent with the original design and typical application in applied research, does not provide an adequate explanation of RSE responses. However, there is no clear agreement about what alternative model is most appropriate—or even a clear rationale for how to test competing interpretations. Three alternative interpretations exist: (a) 2 substantively important trait factors (positive and negative self-esteem), (b) 1 trait factor and ephemeral method artifacts associated with positively or negatively worded items, or (c) 1 trait factor and stable response-style method factors associated with item wording. We have posited 8 alternative models and structural equation model tests based on longitudinal data (4 waves of data across 8 years with a large, representative sample of adolescents). Longitudinal models provide no support for the unidimensional model, undermine support for the 2-factor model, and clearly refute claims that wording effects are ephemeral, but they provide good support for models positing 1 substantive (self-esteem) factor and response-style method factors that are stable over time. This longitudinal methodological approach has not only resolved these long-standing issues in self-esteem research but also has broad applicability to most psychological assessments based on self-reports with a mix of positively and negatively worded items. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A new method to estimate parameters of linear compartmental models using artificial neural networks

At present, the preferred tool for parameter estimation in compartmental analysis is an iterative procedure; weighted nonlinear regression. For a large number of applications, observed data can be fitted to sums of exponentials whose parameters are directly related to the rate constants/coefficients of the compartmental models. Since weighted nonlinear regression often has to be repeated for many different data sets, the process of fitting data from compartmental systems can be very time consuming. Furthermore the minimization routine often converges to a local (as opposed to global) minimum. In this paper, we examine the possibility of using artificial neural networks instead of weighted nonlinear regression in order to estimate model parameters. We train simple feed-forward neural networks to produce as outputs the parameter values of a given model when kinetic data are fed to the networks' input layer. The artificial neural networks produce unbiased estimates and are orders of magnitude faster than regression algorithms. At noise levels typical of many real applications, the neural networks are found to produce lower variance estimates than weighted nonlinear regression in the estimation of parameters from mono- and biexponential models. These results are primarily due to the inability of weighted nonlinear regression to converge. These results establish that artificial neural networks are powerful tools for estimating parameters for simple compartmental models.     

Statistical Analysis on the Cost and Duration of Public Building Projects

A probabilistic model is proposed to predict the risk effects on time and cost of public building projects. The research goal is to utilize a real history data in estimating project cost and duration. The model results can be used to adjust floats and budgets of the planning schedule before project commencement. Statistical regression models and sample tests are developed using real data of 113 public projects. The model outputs can be used by project managers in the planning phase to validate the schedule critical path time and project budget. The comparison of means analysis for project cost and time performance indicated that the sample projects tend to finish over budget and almost on schedule. Regression models were developed to model project cost and time. The regression analysis showed that the project budgeted cost and planned project duration provide a good basis for estimating the cost and duration. The regression model results were validated by estimating the prediction error in percent and through conducting out-of-sample tests. In conclusion, the models were validated at a probability of 95%, at which the proposed models predict the project cost and duration at an error margin of ±0.035% of the actual cost and time.     

Support Vector Machines Approach to HMA Stiffness Prediction

The application of artificial intelligence (AI) techniques to engineering has increased tremendously over the last decade. Support vector machine (SVM) is one efficient AI technique based on statistical learning theory. This paper explores the SVM approach to model the mechanical behavior of hot-mix asphalt (HMA) owing to high degree of complexity and uncertainty inherent in HMA modeling. The dynamic modulus (|E?|), among HMA mechanical property parameters, not only is important for HMA pavement design but also in determining HMA pavement performance associated with pavement response. Previously employed approaches for development of the predictive |E?| models concentrated on multivariate regression analysis of database. In this paper, SVM-based |E?| prediction models were developed using the latest comprehensive |E?| database containing 7,400 data points from 346 HMA mixtures. The developed SVM models were compared with the existing multivariate regression-based |E?| model as well as the artificial neural networks (ANN) based |E?| models developed recently by the writers. The prediction performance of SVM model is better than multivariate regression-based model and comparable to the ANN. Fewer constraints in SVM compared to ANN can make it a promising alternative considering the availability of limited and nonrepresentative data frequently encountered in construction materials characterization.     

Modeling of Temperature–Frequency Correlation Using Combined Principal Component Analysis and Support Vector Regression Technique

A good understanding of environmental effects on structural modal properties is essential for reliable performance of vibration-based damage diagnosis methods. In this paper, a method of combining principal component analysis (PCA) and support vector regression (SVR) technique is proposed for modeling temperature-caused variability of modal frequencies for structures instrumented with long-term monitoring systems. PCA is first applied to extract principal components from the measured temperatures for dimensionality reduction. The predominant feature vectors in conjunction with the measured modal frequencies are then fed into a support vector algorithm to formulate regression models that may take into account thermal inertia effect. The research is focused on proper selection of the hyperparameters to obtain SVR models with good generalization performance. A grid search method with cross validation and a heuristic method are utilized for determining the optimal values of SVR hyperparameters. The proposed method is compared with the method directly using measurement data to train SVR models and the multivariate linear regression (MLR) method through the use of long-term measurement data from a cable-stayed bridge. It is shown that PCA-compressed features make the training and validation of SVR models more efficient in both model accuracy and computational costs, and the formulated SVR model performs much better than the MLR model in generalization performance. When continuously measured data is available, the SVR model formulated taking into account thermal inertia effect can achieve more accurate prediction than that without considering thermal inertia effect.     

A better lemon squeezer? Maximum-likelihood regression with beta-distributed dependent variables.

Uncorrectable skew and heteroscedasticity are among the "lemons" of psychological data, yet many important variables naturally exhibit these properties. For scales with a lower and upper bound, a suitable candidate for models is the beta distribution, which is very flexible and models skew quite well. The authors present maximum-likelihood regression models assuming that the dependent variable is conditionally beta distributed rather than Gaussian. The approach models both means (location) and variances (dispersion) with their own distinct sets of predictors (continuous and/or categorical), thereby modeling heteroscedasticity. The location submodel link function is the logit and thereby analogous to logistic regression, whereas the dispersion submodel is log linear. Real examples show that these models handle the independent observations case readily. The article discusses comparisons between beta regression and alternative techniques, model selection and interpretation, practical estimation, and software. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

New statistical tests of neutrality for DNA samples from a population

The purpose of this paper is to develop statistical tests of the neutral model of evolution against a class of alternative models with the common characteristic of having an excess of mutations that occurred a long time ago or a reduction of recent mutations compared to the neutral model. This class of population genetics models include models for structured populations, models with decreasing effective population size and models of selection and mutation balance. Four statistical tests were proposed in this paper for DNA samples from a population. Two of these tests, one new and another a modification of an existing test, are based on EWENS' sampling formula, and the other two new tests make use of the frequencies of mutations of various classes. Using simulated samples and regression analyses, the critical values of these tests can be computed from regression equations. This approach for computing the critical values of a test was found to be appropriate and quite effective. We examined the powers of these four tests using simulated samples from structured populations, populations with linearly decreasing sizes and models of selection and mutation balance and found that they are more powerful than existing statistical tests of the neutral model of evolution.     

A neural theory of attentive visual search: Interactions of boundary, surface, spatial, and object representations.

Visual search data are given a unified quantitative explanation by a model of how spatial maps in the parietal cortex and object recognition categories in the inferotemporal cortex deploy attentional resources as they reciprocally interact with visual representations in the prestriate cortex. The model visual representations are organized into multiple boundary and surface representations. Visual search in the model is initiated by organizing multiple items that lie within a given boundary or surface representation into a candidate search grouping. These items are compared with object recognition categories to test for matches or mismatches. Mismatches can trigger deeper searches and recursive selection of new groupings until a target object is identified. The model provides an alternative to Feature Integration and Guided Search models. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Poisson Model of Construction Incident Occurrence

Construction incidents are essentially random events because they have a probabilistic component that causes their occurrence to be indeterministic. Thus, as with most random events, one of the best ways to understand and analyze construction incidents is to apply statistical methods and tools. Consequently, this paper presents a statistical framework based on the modified loss causation model (MLCM). Even though the MLCM has been used for the framework, the approach can be readily adapted for other incident causation models. The MLCM is separated into two basic components: random and systematic. The random component is represented by a probability density function (PDF), which has parameters influenced by the systematic component of the MLCM, while the systematic component is represented by the situational variables and quality of the safety management system. In particular, this paper proposes that the PDF can be represented by the Poisson distribution. Besides being a convenient and simple distribution that can be easily used in applications, the Poisson distribution had been used in various industries to model random failures or incidents. The differences in contexts and the undesirable effects of adopting an unrepresentative distribution will require formal analysis to determine the suitability of the Poisson distribution in modeling the random component of construction incident occurrence. Incident records for 14 major projects were used in the analysis. Hypothesis testing using the chi-square goodness-of-fit and dispersion tests shows that the incident occurrences can be modeled as a Poisson process characterized by some mean arrival rate. The paper also presents some applications of the proposed Poisson model to improve construction safety management, focusing on two specific concepts: the Bayesian approach and the partitioned Poisson.     

Expanding test–retest designs to include developmental time-lag components.

Test-retest data can reflect systematic changes over varying intervals of time in a "time-lag" design. This article shows how latent growth models with planned incomplete data can be used to separate psychometric components of developmental interest, including internal consistency reliability, test-practice effects, factor stability, factor growth, and state fluctuation. Practical analyses are proposed using a structural equation model for longitudinal data on multiple groups with different test-retest intervals. This approach is illustrated using 2 sets of data collected from students measured on the Woodcock-Johnson—Revised Memory and Reading scales. The results show how alternative time-lag models can be fitted and interpreted with univariate, bivariate, and multivariate data. Benefits, limitations, and extensions of this structural time-lag approach are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)