Showing papers on "Unit-weighted regression published in 2004"

Regression With Social Data: Modeling Continuous and Limited Response Variables

Abstract: Preface. 1. Introduction to Regression Modeling. 2. Simple Linear Regression. 3. Introduction to Multiple Regression. 4. Multiple Regression with Categorical Predictors: ANOVA and ANCOVA Models. 5. Modeling Nonlinearity. 6. Advanced Issues in Multiple Regression. 7. Regression with a Binary Response. 8. Advanced Topics in Logistic Regression. 9. Truncated and Censored Regression Models. 10. Regression Models for an Event Count. 11. Introduction to Survival Analysis. 12. Multistate, Multiepisode, and Interval-Censored Models in Survival Analysis. Appendix A: Mathematics Tutorials. Appendix B: Answers to Selected Exercises. References. Index.

Applied Statistics: Analysis of Variance and Regression

Abstract: the textbook, online testing, diagnosis, tutorials, and grade books. A separate Technology Manual and Workbook provides examples requiring technology, such as the TI-83+ graphing calculator, Excel, MINITAB, and STATDISK. The workbook comes with STATDISK and DDXL software. Supplements for instructors include an Instructor’s Guide and Solution Manual by David Lund; a computerized test generator, TestGen-Eq, for Windows and Macintosh; and three sets of tests in the Test Bank. Supplements for students include a Student Solutions Manual by David Lund and tutoring through the AWL Math Tutor Center. An introductory book must carefully consider its scope and make difficult decisions to be successful with its intended audience. The authors have done well in this regard. The material provides sufficient material to develop a statistically literate citizen in the modern world. It achieves this result largely by addressing the statistical information and concepts, such as “margin of error,” that are encountered everywhere in our normal surroundings and maintaining their context, instead of presenting an isolated statistical concept as abstract lessons. A common form of statistics that is not commonly found in other text books is index numbers, which this book addresses at length. The book is written in a clear and compelling style. It uses ample visual aids and graphic designs to make the lessons engaging and to help organize new information for the student. Its style should appeal to its intended audience, but it might be criticized by instructors or students who appreciate a more traditional mathematics book. There is an overall rewarding sense of completeness and clarity at the conclusion of each presentation. I found no glaring mistakes to detract from the flow of the presentation. Every topic is introduced in the context of at least one real-life example or case from history or the present to learn from and practice. Each section includes questions and extra problems in the form of “Projects for the Web and Beyond.” Each chapter ends with a strong example in the “Focus on . . .” highlight. Topics include medicine, education, ethics, politics, and the environment. More theoretical or advanced information is called out into the book margin as a “Technical Note.” There is a good deal of material about important statistical plots, what kind of plot to use with particular data, and specific pitfalls and misuse. Other good points include a strong section on the distinction between correlation and causality. Just a couple of minor bad points include a rule of thumb for computing the sample standard deviation from the sample range and computing “low and high values” (sample mean plus or minus two sample standard deviations) without interpretation or reference to a distribution. Statistical Reasoning for Everyday Life should be appreciated by its intended audience and popular. It could also be used by business and technical professionals with little previous statistics training for self-study or refresher.

A Note on the Mixed Geographically Weighted Regression Model

Abstract: . A mixed, geographically weighted regression (GWR) model is useful in thesituation where certain explanatory variables influencing the response are global whileothers are local. Undoubtedly, how to identify these two types of the explanatory variablesisessentialforbuildingsuchamodel.Nevertheless,Itseemsthattherehasnotbeenaformalway to achieve this task. Based on some work on the GWR technique and the distributiontheory of quadratic forms in normal variables, a statistical test approach is suggested here toidentify a mixed GWR model. Then, this note mainly focuses on simulation studies toexamine the performance of the test and to provide some guidelines for performing the testin practice. The simulation studies demonstrate that the test works quite well and provides afeasible way to choose an appropriate mixed GWR model for a given data set. 1. INTRODUCTIONGeographically weighted regression (GWR) (Brunsdon, Fotheringham,and Charlton, 1996, 1998; Fotheringham, Charlton, and Brunsdon, 1997a,1997b) is a useful technique for locally modeling of spatial relationships bycalibrating a varying coefficient regression model of the formy

Applied Regression Analysis: A Second Course in Business and Economic Statistics

Abstract: 1. An Introduction to Regression Analysis. 2. Review of Basic Statistical Concepts. Introduction / Descriptive Statistics / Discrete Random Variables and Probability Distributions / The Normal Distribution / Populations, Samples, and Sampling Distributions / Estimating a Population Mean / Hypothesis Tests About a Population Mean / Estimating the Difference Between Two Population Means / Hypothesis Tests About the Difference Between Two Population Means. 3. Simple Regression Analysis. Using Simple Regression to Describe a Linear Relationship / Examples of Regression as a Descriptive Technique / Inferences from a Simple Regression Analysis / Assessing the Fit of the Regression Line / Prediction or Forecasting with a Simple Linear Regression Equation. Fitting a Linear Trend to Time-Series Data / Some Cautions in Interpreting Regression Results. 4. Multiple Regression Analysis. Using Multiple Regression to Describe a Linear Relationship / Inferences from a Multiple Regression Analysis / Assessing the Fit of the Regression Line / Comparing Two Regression Models / Prediction with a Multiple Regression Equation / Multicollinearity: A Potential Problem in Multiple Regression / Lagged Variables as Explanatory Variables in Time-Series Regression. 5. Fitting Curves to Data. Introduction / Fitting Curvilinear Relationships. 6. Assessing the Assumptions of the Regression Model. Introduction. Assumptions of the Multiple Linear Regression Model / The Regression Residuals / Assessing the Assumption That the Relationship is Linear / Assessing the Assumption That the Variance Around the Regression Line is Constant / Assessing the Assumption That the Disturbances are Normally Distributed / Influential observations / Assessing the Influence That the Disturbances are Independent. 7. Using Indicator and Interaction Variables. Using and Interpreting Indicator Variables / Interaction Variables / Seasonal Effects in Time-Series Regression. 8. Variable Selection. Introduction. All Possible Regressions. Other Variable Selection Techniques / Which Variable Selection Procedure is Best? 9. An Introduction to Analysis of Variance. One-Way Analysis of Variance. Analysis of Variance Using a Randomized Block Design / Two-Way Analysis of Variance / Analysis of Covariance. 10. Qualitative Dependent Variables: An Introduction to Discriminant Analysis and Logistic Regression. Introduction. Discriminant Analysis / Logistic Regression. 11. Forecasting Methods for Time-Series Data. Introduction / Naive Forecasts / Measuring Forecast Accuracy / Moving Averages / Exponential Smoothing / Decomposition. APPENDICES. A: Summation Notation. B: Statistical Tables. C: A Brief Introduction to MINITAB, Microsoft Excel, and SAS. D: Matrices and their Application to Regression Analysis. E: Solutions to Selected Odd-Numbered Exercises. References / Index.

Outlier Detecting in Fuzzy Switching Regression Models

Abstract: Fuzzy switching regression models have been extensively used in economics and data mining research. We present a new algorithm named FCWRM (Fuzzy C Weighted Regression Model) to detect the outliers in fuzzy switching regression models while preserving the merits of FCRM algorithm proposed by Hathaway. The theoretic analysis shows that FCWRM can converge to a local minimum of the object function. Several numeric examples demonstrate the effectiveness of algorithm FCWRM.

Outlier detecting in fuzzy switching regression models

Abstract: Fuzzy switching regression models have been extensively used in economics and data mining research. We present a new algorithm named FCWRM (Fuzzy C Weighted Regression Model) to detect the outliers in fuzzy switching regression models while preserving the merits of FCRM algorithm proposed by Hathaway. The theoretic analysis shows that FCWRM can converge to a local minimum of the object function. Several numeric examples demonstrate the effectiveness of algorithm FCWRM.

Enhance-synergism and suppression effects in multiple regression

Abstract: Relations between pairwise correlations and the coefficient of multiple determination in regression analysis are considered. The conditions for the occurrence of enhance-synergism and suppression effects when multiple determination becomes bigger than the total of squared correlations of the dependent variable with the regressors are discussed. It is shown that such effects can occur just for stochastic relations among the variables that have non-transitive signs of pairwise correlations. Consideration of these problems facilitates better understanding the properties of regression.

On fuzzy clustering based regression models

Abstract: We have proposed a fuzzy cluster loading [7] which can show the relationship between the obtained fuzzy clusters and the variables in order to interpret the obtained fuzzy clusters. Moreover, we have proposed a weighted regression model using a fuzzy clustering result obtained by the classification of the data with respect to explanatory variables. [8] These models are closely related with the conventional geographically weighted regression model [2] and the fuzzy c-regression model. [5] So, this paper discusses the difference and the relation of these models from the view point of the difference of the estimates of the regression coefficients and the assumption of the errors. Several numerical examples show the difference and the better performance of the proposed models.

Ridge Regression as an Alternative to Ordinary Least Squares: Improving Prediction Accuracy and the Interpretation of Beta Weights

Abstract: This article looked at non-experimental data via an ordinary least squares (OLS) model and compared its results to ridge regression models in terms of crossvalidation predictor weighting precision when using fixed and random predictor cases and small and large p/n ratio models. A majority of the time with two random predictor cases, ridge regression accuracy was superior to OLS in estimating beta weights. Thus, ridge regression was very useful under this condition. However, when the fixed predictor case was reviewed, OLS was much more precise at estimating predictor weights than the ridge techniques regardless of the p/n ratio. In determining the cross validation accuracy of the ridge estimated weights in respect to the OLS estimated weights, ridge models were superior for improving the accuracy of model prediction.