Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
482,177
result(s) for
"Regression"
Sort by:
Regression analysis : a practical introduction
\"With the rise of \"big data\", there is an increasing demand to learn the skills needed to undertake sound quantitative analysis without requiring students to spend too much time on high-level math and proofs. This book provides an efficient alternative approach, with more time devoted to the practical aspects of regression analysis and how to recognise the most common pitfalls. By doing so, the book will better prepare readers for conducting, interpreting, and assessing regression analyses, while simultaneously making the material simpler and more enjoyable to learn. Logical and practical in approach, Regression Analysis teaches: (1) the tools for conducting regressions; (2) the concepts needed to design optimal regression models (based on avoiding the pitfalls); and (3) the proper interpretations of regressions. Furthermore, this book emphasizes honesty in research, with a prevalent lesson being that statistical significance is not the goal of research. This book is an ideal introduction to regression analysis for anyone learning quantitative methods in the social sciences, business, medicine, and data analytics. It will also appeal to researchers and academics looking to better understand what regressions do, what their limitations are, and what they can tell us. This will be the most engaging book on regression analysis (or Econometrics) you will ever read!\"-- Provided by publisher.
Book
Unconditional Quantile Regressions
We propose a new regression method to evaluate the impact of changes in the distribution of the explanatory variables on quantiles of the unconditional (marginal) distribution of an outcome variable. The proposed method consists of running a regression of the (recentered) influence function (RIF) of the unconditional quantile on the explanatory variables. The influence function, a widely used tool in robust estimation, is easily computed for quantiles, as well as for other distributional statistics. Our approach, thus, can be readily generalized to other distributional statistics.
Journal Article
Handbook of regression analysis
\"Written by two established experts in the field, the purpose of this handbook is to provide a practical, one-stop reference on regression analysis. The focus is on the tools that both practitioners and researchers use in real life. It is intended to be a comprehensive collection of the theory, methods, and applications of the subject matter, but it is deliberately written at an accessible level. The handbook will provide a quick and convenient reference or \"refresher\" on ideas and methods that are useful for the accurate analysis of data and its resulting interpretations. Students can use the book as an introduction to and/or summary of key concepts in regression and related course work (such as linear, nonlinear, and nonparametric regressions). Plentiful references are supplied for the more motivated readers. Theory is presented when necessary, and always supplemented by hands-on examples. Software routines are available via an author-maintained web site\"-- Provided by publisher.
Book
Machine-Learning-Algorithm to predict the High-Performance concrete compressive strength using multiple data
PurposeThe compressive strength of concrete depends on many interdependent parameters; its exact prediction is not that simple because of complex processes involved in strength development. This study aims to predict the compressive strength of normal concrete and high-performance concrete using four datasets.Design/methodology/approachIn this paper, five established individual Machine Learning (ML) regression models have been compared: Decision Regression Tree, Random Forest Regression, Lasso Regression, Ridge Regression and Multiple-Linear regression. Four datasets were studied, two of which are previous research datasets, and two datasets are from the sophisticated lab using five established individual ML regression models.FindingsThe five statistical indicators like coefficient of determination (R2), mean absolute error, root mean squared error, Nash–Sutcliffe efficiency and mean absolute percentage error have been used to compare the performance of the models. The models are further compared using statistical indicators with previous studies. Lastly, to understand the variable effect of the predictor, the sensitivity and parametric analysis were carried out to find the performance of the variable.Originality/valueThe findings of this paper will allow readers to understand the factors involved in identifying the machine learning models and concrete datasets. In so doing, we hope that this research advances the toolset needed to predict compressive strength.
Journal Article
Bayesian Inference for Logistic Models Using Pólya–Gamma Latent Variables
We propose a new data-augmentation strategy for fully Bayesian inference in models with binomial likelihoods. The approach appeals to a new class of Pólya–Gamma distributions, which are constructed in detail. A variety of examples are presented to show the versatility of the method, including logistic regression, negative binomial regression, nonlinear mixed-effect models, and spatial models for count data. In each case, our data-augmentation strategy leads to simple, effective methods for posterior inference that (1) circumvent the need for analytic approximations, numerical integration, or Metropolis–Hastings; and (2) outperform other known data-augmentation strategies, both in ease of use and in computational efficiency. All methods, including an efficient sampler for the Pólya–Gamma distribution, are implemented in the R package BayesLogit . Supplementary materials for this article are available online.
Journal Article
Mostly harmless econometrics : an empiricist's companion
In addition to econometric essentials, this book covers important new extensions as well as how to get standard errors right. The authors explain why fancier econometric techniques are typically unnecessary and even dangerous.
Book
P286 Is there diurnal or seasonal variation in the 6-minute walk test distance in patients with pulmonary vascular disease?
IntroductionThe 2002 American Thoracic Society (ATS) 6-Minute Walk Test (6MWT) guidelines advise ‘repeat testing should be performed about the same time of day to minimize intraday variability’. It is currently unknown if there is diurnal and/or seasonal variation in the 6MWT in pulmonary vascular disease (PVD).AimsTo investigate the association between 6-Minute Walk Distance (6MWD), time of day and season of testing.MethodRetrospective 6MWT data was collected from 2019–2025. The 6MWT was conducted in accordance with ATS 6MWT guidelines (2002). Independent sample t-test was performed to determine morning vs afternoon differences in 6MWD. One way ANOVA was performed to determine seasonal differences. Time of day and seasonal effects were further investigated using multivariate linear regression analyses, adjusted for age, sex, BMI and WHO functional class. Time was modelled as a continuous variable (per 1 hour increment) and as a binary predictor of morning (08:00–12:00 am) vs afternoon (12–18:00 pm).ResultsThe cohort consisted of 1618 patients, WHO PH Groups 1 (n= 322), 2 (n=72), 3 (n= 51), 4 (n = 946), and chronic thromboembolic pulmonary disease (n= 227). 6MWD was significantly higher in the morning (363 ± 134 meters; n=684) than in the afternoon (333 ± 144 meters; n=865) (p<0.001). 6MWD was not significantly impacted by season (p=0.75). Multivariable linear regression analyses demonstrated that the 6MWD decreased by 4.1 meters (95% CI 1.57–6.64) (p=0.002) for every hour increment in the time of the day, and by 16.34 meters (95% CI 5.72–29.95) (p=0.0003) in the afternoon compared to the morning.ConclusionThe time of day was found to impact 6MWD, with patients with PVD achieving higher 6MWD in the morning. 6MWTs should be conducted at the same time to eliminate this possible fatiguing effect and optimise interpretation of intertest differences in clinical practice and trials. Prospective studies are needed with repeated measurements in the same individuals at different times of the day to validate these findings.
Journal Article
Regression models for categorical, count, and related variables : an applied approach
\"Social science and behavioral science students and researchers are often confronted with data that are categorical, count a phenomenon, or have been collected over time. Sociologists examining the likelihood of interracial marriage, political scientists studying voting behavior, criminologists counting the number of offenses people commit, health scientists studying the number of suicides across neighborhoods, and psychologists modeling mental health treatment success are all interested in outcomes that are not continuous. Instead, they must measure and analyze these events and phenomena in a discrete manner. This book provides an introduction and overview of several statistical models designed for these types of outcomes--all presented under the assumption that the reader has only a good working knowledge of elementary algebra and has taken introductory statistics and linear regression analysis. Numerous examples from the social sciences demonstrate the practical applications of these models. The chapters address logistic and probit models, including those designed for ordinal and nominal variables, regular and zero-inflated Poisson and negative binomial models, event history models, models for longitudinal data, multilevel models, and data reduction techniques such as principal components and factor analysis. Each chapter discusses how to utilize the models and test their assumptions with the statistical software Stata, and also includes exercise sets so readers can practice using these techniques. Appendices show how to estimate the models in SAS, SPSS, and R; provide a review of regression assumptions using simulations; and discuss missing data. A companion website includes downloadable versions of all the data sets used in the book\"--Provided by publisher.
Book
Simultaneous Estimation and Variable Selection for Interval-Censored Data With Broken Adaptive Ridge Regression
The simultaneous estimation and variable selection for Cox model has been discussed by several authors when one observes right-censored failure time data. However, there does not seem to exist an established procedure for interval-censored data, a more general and complex type of failure time data, except two parametric procedures. To address this, we propose a broken adaptive ridge (BAR) regression procedure that combines the strengths of the quadratic regularization and the adaptive weighted bridge shrinkage. In particular, the method allows for the number of covariates to be diverging with the sample size. Under some weak regularity conditions, unlike most of the existing variable selection methods, we establish both the oracle property and the grouping effect of the proposed BAR procedure. An extensive simulation study is conducted and indicates that the proposed approach works well in practical situations and deals with the collinearity problem better than the other oracle-like methods. An application is also provided.
Journal Article