Showing papers on "Random effects model published in 1998"

Statistical analysis of repeated measures data using SAS procedures.

Abstract: Mixed linear models were developed by animal breeders to evaluate genetic potential of bulls. Application of mixed models has recently spread to all areas of research, spurred by availability of advanced computer software. Previously, mixed model analyses were implemented by adapting fixed-effect methods to models with random effects. This imposed limitations on applicability because the covariance structure was not modeled. This is the case with PROC GLM in the SAS® System. Recent versions of the SAS System include PROC MIXED. This procedure implements random effects in the statistical model and permits modeling the covariance structure of the data. Thereby, PROC MIXED can compute efficient estimates of fixed effects and valid standard errors of the estimates. Modeling the covariance structure is especially important for analysis of repeated measures data because measurements taken close in time are potentially more highly correlated than those taken far apart in time.

Fixed- and random-effects models in meta-analysis.

Abstract: There are 2 families of statistical procedures in meta-analysis: fixed- and randomeffects procedures. They were developed for somewhat different inference goals: making inferences about the effect parameters in the studies that have been observed versus making inferences about the distribution of effect parameters in a population of studies from a random sample of studies. The authors evaluate the performance of confidence intervals and hypothesis tests when each type of statistical procedure is used for each type of inference and confirm that each procedure is best for making the kind of inference for which it was designed. Conditionally random-effects procedures (a hybrid type) are shown to have properties in between those of fixed- and random-effects procedures. The use of quantitative methods to summarize the results of several empirical research studies, or metaanalysis, is now widely used in psychology, medicine, and the social sciences. Meta-analysis usually involves describing the results of each study by means of a numerical index (an estimate of effect size, such as a correlation coefficient, a standardized mean difference, or an odds ratio) and then combining these estimates across studies to obtain a summary. Two somewhat different statistical models have been developed for inference about average effect size from a collection of studies, called the fixed-effects and random-effects models. (A third alternative, the mixedeffects model, arises in conjunction with analyses involving study-level covariates or moderator variables, which we do not consider in this article; see Hedges, 1992.) Fixed-effects models treat the effect-size parameters as fixed but unknown constants to be estimated and usually (but not necessarily) are used in conjunction with assumptions about the homogeneity of effect parameters (see, e.g., Hedges, 1982; Rosenthal & Rubin, 1982). Random-effects models treat the effectsize parameters as if they were a random sample from

Model choice: A minimum posterior predictive loss approach

Abstract: SUMMARY Model choice is a fundamental and much discussed activity in the analysis of datasets. Nonnested hierarchical models introducing random effects may not be handled by classical methods. Bayesian approaches using predictive distributions can be used though the formal solution, which includes Bayes factors as a special case, can be criticised. We propose a predictive criterion where the goal is good prediction of a replicate of the observed data but tempered by fidelity to the observed values. We obtain this criterion by minimising posterior loss for a given model and then, for models under consideration, selecting the one which minimises this criterion. For a broad range of losses, the criterion emerges as a form partitioned into a goodness-of-fit term and a penalty term. We illustrate its performance with an application to a large dataset involving residential property transactions.

Detecting and describing heterogeneity in meta-analysis

Abstract: The investigation of heterogeneity is a crucial part of any meta-analysis. While it has been stated that the test for heterogeneity has low power, this has not been well quantified. Moreover the assumptions of normality implicit in the standard methods of meta-analysis are often not scrutinized in practice. Here we simulate how the power of the test for heterogeneity depends on the number of studies included, the total information (that is total weight or inverse variance) available and the distribution of weights among the different studies. We show that the power increases with the total information available rather than simply the number of studies, and that it is substantially lowered if, as is quite common in practice, one study comprises a large proportion of the total information. We also describe normal plots that are useful in assessing whether the data conform to a fixed effect or random effects model, together with appropriate tests, and give an application to the analysis of a multi-centre trial of blood pressure reduction. We conclude that the test of heterogeneity should not be the sole determinant of model choice in meta-analysis, and inspection of relevant normal plots, as well as clinical insight, may be more relevant to both the investigation and modelling of heterogeneity.

Extending the simple linear regression model to account for correlated responses: an introduction to generalized estimating equations and multi-level mixed modelling.

Abstract: Much of the research in epidemiology and clinical science is based upon longitudinal designs which involve repeated measurements of a variable of interest in each of a series of individuals. Such designs can be very powerful, both statistically and scientifically, because they enable one to study changes within individual subjects over time or under varied conditions. However, this power arises because the repeated measurements tend to be correlated with one another, and this must be taken into proper account at the time of analysis or misleading conclusions may result. Recent advances in statistical theory and in software development mean that studies based upon such designs can now be analysed more easily, in a valid yet flexible manner, using a variety of approaches which include the use of generalized estimating equations, and mixed models which incorporate random effects. This paper provides a particularly simple illustration of the use of these two approaches, taking as a practical example the analysis of a study which examined the response of portable peak expiratory flow meters to changes in true peak expiratory flow in 12 children with asthma. The paper takes the reader through the relevant practicalities of model fitting, interpretation and criticism and demonstrates that, in a simple case such as this, analyses based upon these model-based approaches produce reassuringly similar inferences to standard analyses based upon more conventional methods.

Size is everything--large amounts of information are needed to overcome random effects in estimating direction and magnitude of treatment effects.

Abstract: Abstract Variability in patients' response to interventions in pain and other clinical settings is large. Many explanations such as trial methods, environment or culture have been proposed, but this paper sets out to show that the main cause of the variability may be random chance, and that if trials are small their estimate of magnitude of effect may be incorrect, simply because of the random play of chance. This is highly relevant to the questions of ‘How large do trials have to be for statistical accuracy?’ and ‘How large do trials have to be for their results to be clinically valid?’ The true underlying control event rate (CER) and experimental event rate (EER) were determined from single‐dose acute pain analgesic trials in over 5000 patients. Trial group size required to obtain statistically significant and clinically relevant (0.95 probability of number‐needed‐to‐treat within ±0.5 of its true value) results were computed using these values. Ten thousand trials using these CER and EER values were simulated using varying group sizes to investigate the variation due to random chance alone. Most common analgesics have EERs in the range 0.4–0.6 and CER of about 0.19. With such efficacy, to have a 90% chance of obtaining a statistically significant result in the correct direction requires group sizes in the range 30–60. For clinical relevance nearly 500 patients are required in each group. Only with an extremely effective drug (EER>0.8) will we be reasonably sure of obtaining a clinically relevant NNT with commonly used group sizes of around 40 patients per treatment arm. The simulated trials showed substantial variation in CER and EER, with the probability of obtaining the correct values improving as group size increased. We contend that much of the variability in control and experimental event rates is due to random chance alone. Single small trials are unlikely to be correct. If we want to be sure of getting correct (clinically relevant) results in clinical trials we must study more patients. Credible estimates of clinical efficacy are only likely to come from large trials or from pooling multiple trials of conventional (small) size.

Semiparametric Stochastic Mixed Models for Longitudinal Data

Abstract: We consider inference for a semiparametric stochastic mixed model for longitudinal data. This model uses parametric fixed effects to represent the covariate effects and an arbitrary smooth function to model the time effect and accounts for the within-subject correlation using random effects and a stationary or nonstationary stochastic process. We derive maximum penalized likelihood estimators of the regression coefficients and the nonparametric function. The resulting estimator of the nonparametric function is a smoothing spline. We propose and compare frequentist inference and Bayesian inference on these model components. We use restricted maximum likelihood to estimate the smoothing parameter and the variance components simultaneously. We show that estimation of all model components of interest can proceed by fitting a modified linear mixed model. We illustrate the proposed method by analyzing a hormone dataset and evaluate its performance through simulations.

Modelling risk from a disease in time and space.

Abstract: This paper combines existing models for longitudinal and spatial data in a hierarchical Bayesian framework, with particular emphasis on the role of time- and space-varying covariate effects. Data analysis is implemented via Markov chain Monte Carlo methods. The methodology is illustrated by a tentative re-analysis of Ohio lung cancer data 1968-1988. Two approaches that adjust for unmeasured spatial covariates, particularly tobacco consumption, are described. The first includes random effects in the model to account for unobserved heterogeneity; the second adds a simple urbanization measure as a surrogate for smoking behaviour. The Ohio data set has been of particular interest because of the suggestion that a nuclear facility in the southwest of the state may have caused increased levels of lung cancer there. However, we contend here that the data are inadequate for a proper investigation of this issue.

Meta-analysis of multiple outcomes by regression with random effects.

Abstract: Earlier work showed how to perform fixed-effects meta-analysis of studies or trials when each provides results on more than one outcome per patient and these multiple outcomes are correlated. That fixed-effects generalized-least-squares approach analyzes the multiple outcomes jointly within a single model, and it can include covariates, such as duration of therapy or quality of trial, that may explain observed heterogeneity of results among the trials. Sometimes the covariates explain all the heterogeneity, and the fixed-effects regression model is appropriate. However, unexplained heterogeneity may often remain, even after taking into account known or suspected covariates. Because fixed-effects models do not make allowance for this remaining unexplained heterogeneity, the potential exists for bias in estimated coefficients, standard errors and p-values. We propose two random-effects approaches for the regression meta-analysis of multiple correlated outcomes. We compare their use with fixed-effects models and with separate-outcomes models in a meta-analysis of periodontal clinical trials. A simulation study shows the advantages of the random-effects approach. These methods also facilitate meta-analysis of trials that compare more than two treatments.

Evaluating Median Crossover Likelihoods with Clustered Accident Counts: An Empirical Inquiry Using the Random Effects Negative Binomial Model

Abstract: Insights into plausible methodological frameworks specifically with respect to two key issues—(1) mathematical formulation of the underlying process affecting median crossover accidents and (2) the factors affecting median crossover frequencies in Washington State—are provided in this study. Random effects negative binomial (RENB) and the cross-sectional negative binomial (NB) models are examined. The specification comparisons indicate benefits from using the RENB model only when spatial and temporal effects are totally unobserved. When spatial and temporal effects are explicitly included, the NB model is statistically adequate, while the RENB model appears to lose its distributional advantage. Such findings might be artifacts of the median crossover accident dataset used in this study. While the NB model appears to be the superior model in the present case of median crossover accidents, the marginally inferior performance of the RENB model warrants further examination through application to regular accid...

A Hierarchical Bayes Model of Primary and Secondary Demand

Abstract: Product design, pricing policies, and promotional activities influence the primary and secondary demand for goods and services. Brand managers need to develop an understanding of the relationships between marketing mix decisions and consumer decisions of whether to purchase in the product category, which brand to buy, and how much to consume. Knowledge about factors most effective in influencing primary and secondary demand of a product allows firms to grow by enhancing their market share as well their market size. The purpose of this paper is to develop an individual level model that allows an investigation of both the primary and secondary aspects of consumer demand. Unlike models of only primary demand or only secondary demand, this more comprehensive model offers the opportunity to identify changes in product features that will result in the greatest increase in demand. It also offers the opportunity to differentially target consumer segments depending upon whether consumers are most likely to enter the market, increase their consumption level, or switch brands. In the proposed hierarchical Bayes model, an integrative framework that jointly models the discrete choice and continuous quantity components of consumer decision is employed instead of treating the two as independent. The model includes parameters that capture individual specific reservation value, attribute preference, and expenditure sensitivity. The model development is based upon the microeconomic theory of utility maximization. Heterogeneity in model parameters across the sample is captured by using a random effects specification guided by the underlying microeconomic model. This requires that some of the effects are strictly positive. This is accommodated through the use of a gamma distribution of heterogeneity for some of the parameters. A normal distribution of heterogeneity is used for the remaining parameters. Gibbs sampling is used to estimate the model. The key methodological contribution of this paper is that we show how to specify a hierarchical Bayes continuous random effects model that integrates consumer choice and quantity decisions such that individual-level parameters can be estimated. Individual level estimates are desirable because insights into primary demand involve nonlinear functions of model parameters. For example, consumers not in the market are those whose utilities for the choice alternatives fall below some reservation value. The proposed methodology yields individual specific estimates of reservation values and expenditure sensitivity, which allow assessment of the origins of demand other than the switch ing behavior of consumers. The methodology can also be used to help identify changes in product features most likely to bring new customers into a market. Our work differs from previous research in this area as we lay the framework needed to obtain individual-level parameter estimates in a continuous random effects model that integrates choice and quantity. The methodology is demonstrated with survey data collected about consumer preferences and consumption for a food item. For the data available, a large response heterogeneity was observed across all model parameters. In spite of limited data available at the individual level, a majority of the individual level estimates were found to be significant. Predictive tests demonstrated the superiority of the proposed model over existing latent class and aggregate models. Particularly, significant gains in predictive accuracy were observed for the "no-buy" behavior of the respondents. These gains demonstrate that by structurally linking the choice and quantity models results in a more accurate characterization of the market than existing finite mixture approaches that model choice and quantity independently. We show that our joint model makes more efficient use of the available data and results in better parameter estimates than those that assume independence. Finally, the individual level demand analysis is illustrated through a simple example involving a $1.00 price cut. We demonstrate practical usefulness of the model for targeting by developing the demographic, attitudinal, and behavioral profiles of consumer groups most likely to increase consumption, enter the market, or switch brands because of a price cut decision.

Nonparametric Regression Analysis of Longitudinal Data

Abstract: Nonparametric methods are developed for estimating the dose effect when a response consists of correlated observations over time measured in a dose–response experiment. The methods can also be applied to data collected from a completely randomized design experiment. Methods are developed for the detection and description of the effects of dose, time, and their interaction. The methods allow for individual variation in the timing and number of observations. A generalization allowing baseline covariates to be incorporated is addressed. These results may be used in an exploratory fashion in the process of building a random-effects model for longitudinal data.

A Semiparametric Bayesian Approach to the Random Effects Model

Abstract: In longitudinal random effects models, the random effects are typically assumed to have a normal distribution in both Bayesian and classical models. We provide a Bayesian model that allows the random effects to have a nonparametric prior distribution. We propose a Dirichlet process prior for the distribution of the random effects; computation is made possible by the Gibbs sampler. An example using marker data from an AIDS study is given to illustrate the methodology.

Mixed effects smoothing spline analysis of variance

Abstract: We propose a general family of nonparametric mixed effects models. Smoothing splines are used to model the fixed effects and are estimated by maximizing the penalized likelihood function. The random effects are generic and are modelled parametrically by assuming that the covariance function depends on a parsimonious set of parameters. These parameters and the smoothing parameter are estimated simultaneously by the generalized maximum likelihood method. We derive a connection between a nonparametric mixed effects model and a linear mixed effects model. This connection suggests a way of fitting a nonparametric mixed effects model by using existing programs. The classical two-way mixed models and growth curve models are used as examples to demonstrate how to use smoothing spline analysis-of-variance decompositions to build nonparametric mixed effects models. Similarly to the classical analysis of variance, components of these nonparametric mixed effects models can be interpreted as main effects and interactions. The penalized likelihood estimates of the fixed effects in a two-way mixed model are extensions of James-Stein shrinkage estimates to correlated observations. In an example three nested nonparametric mixed effects models are fitted to a longitudinal data set.

Standard Errors of Prediction in Generalized Linear Mixed Models

Abstract: The unconditional mean squared error of prediction (UMSEP) is widely used as a measure of prediction variance for inferences concerning linear combinations of fixed and random effects in the classical normal theory mixed model. But the UMSEP is inappropriate for generalized linear mixed models where the conditional variance of the random effects depends on the data. When the random effects describe variation between independent small domains and domain-specific prediction is of interest, we propose a conditional mean squared error of prediction (CMSEP) as a general measure of prediction variance. The CMSEP is shown to be the sum of the conditional variance and a positive correction that accounts for the sampling variability of parameter estimates. We derive a second-order-correct estimate of the CMSEP that consists of three components: (a) a plug-in estimate of the conditional variance, (b) a plug-in estimate of a Taylor series approximation to the correction term, and (c) a bootstrap estimate of...

Mixed effects logistic regression models for longitudinal binary response data with informative drop-out.

Abstract: A shared parameter model with logistic link is presented for longitudinal binary response data to accommodate informative drop-out. The model consists of observed longitudinal and missing response components that share random effects parameters. To our knowledge, this is the first presentation of such a model for longitudinal binary response data. Comparisons are made to an approximate conditional logit model in terms of a clinical trial dataset and simulations. The naive mixed effects logit model that does not account for informative drop-out is also compared. The simulation-based differences among the models with respect to coverage of confidence intervals, bias, and mean squared error (MSE) depend on at least two factors: whether an effect is a between- or within-subject effect and the amount of between-subject variation as exhibited by variance components of the random effects distributions. When the shared parameter model holds, the approximate conditional model provides confidence intervals with good coverage for within-cluster factors but not for between-cluster factors. The converse is true for the naive model. Under a different drop-out mechanism, when the probability of drop-out is dependent only on the current unobserved observation, all three models behave similarly by providing between-subject confidence intervals with good coverage and comparable MSE and bias but poor within-subject confidence intervals, MSE, and bias. The naive model does more poorly with respect to the within-subject effects than do the shared parameter and approximate conditional models. The data analysis, which entails a comparison of two pain relievers and a placebo with respect to pain relief, conforms to the simulation results based on the shared parameter model but not on the simulation based on the outcome-driven drop-out process. This comparison between the data analysis and simulation results may provide evidence that the shared parameter model holds for the pain data.

Bias Analysis and SIMEX Approach in Generalized Linear Mixed Measurement Error Models

Abstract: We consider generalized linear mixed models (GLMMs) for clustered data when one of the predictors is measured with error. When the measurement error is additive and normally distributed and the error-prone predictor is itself normally distributed, we show that the observed data also follow a GLMM but with a different fixed effects structure from the original model, a different and more complex random effects structure, and restrictions on the parameters. This characterization enables us to compute the biases that result in common GLMMs when one ignores measurement error. For instance, in one common situation the biases in parameter estimates become larger as the number of observations within a cluster increases, both for regression coefficients and for variance components. Parameter estimation is described using the SIMEX method, a relatively new functional method that makes no assumptions about the structure of the unobservable predictors. Simulations and an example illustrate the results.

A general framework for random effects survival analysis in the cox proportional hazards setting

Abstract: The use of random effects modeling in statistics has increased greatly in recent years. The introduction of such modeling into event-time analysis has proceeded more slowly, however. Previously, random effects models for survival data have either required assumptions regarding the form of the baseline hazard function or restrictions on the classes of models that can be fit. In this paper, we develop a method of random effect analysis of survival data, the hierarchical Cox model, that is an extension of Cox's original formulation in that the baseline hazard function remains unspecified. This method also allows an arbitrary distribution for the random effects. We accomplish this using Markov chain Monte Carlo methods in a Bayesian setting. The method is illustrated with three models for a dataset with times to multiple occurrences of mammory tumors for 48 rats treated with a carcinogen and then randomized to either treatment or control. This analysis is more satisfying than standard approaches, such as studying the first event for each subject, which does not fully use the data, or assuming independence, which in this case would overestimate the precision.

A comparison of the random-effects pattern mixture model with last-observation-carried-forward(locf) analysis in longitudinal clinical trials with dropouts

Abstract: The last-observation-carried-forward imputation method is commonly used for imputting data missing due to dropouts in longitudinal clinical trials. The method assumes that outcome remains constant at the last observed value after dropout, which is unlikely in many clinical trials. Recently, random-effects regression models have become popular for analysis of longitudinal clinical trial data with dropouts. However, inference obtained from random-effects regression models is valid when the missing-at-random dropout process is present. The random-effects pattern-mixture model, on the other hand, provides an approach that is valid under more general missingness mechanisms. In this article we describe the use of random-effects pattern-mixture models under different patterns for dropouts. First, subjects are divided into groups depending on their missing-data patterns, and then model parameters are estimated for each pattern. Finally, overall estimates are obtained by averaging over the missing-data patterns and corresponding standard errors are obtained using the delta method. A typical longitudinal clinical trial data set is used to illustrate and compare the above methods of data analyses in the presence of missing data due to dropouts.

An alternative method for meta-analysis

Abstract: In many fields of applications, test statistics are obtained by combining estimates from several experiments, studies or centres of a multicentre trial. The commonly used test procedure to judge the evidence of a common overall effect can result in a considerable overestimation of the significance level, leading to a high rate of too liberal decisions. An alternative test statistic is presented and a better approximating test distribution is derived. Explicitely discussed are the methods in the unbalanced heteroscedastic, way random ANOVA model and for the probability difference method including interaction treatment by centres. Numerical results are presented by simulation studies.

Fast EM-type implementations for mixed effects models

Abstract: The mixed effects model, in its various forms, is a common model in applied statistics. A useful strategy for fitting this model implements EM-type algorithms by treating the random effects as missing data. Such implementations, however, can be painfully slow when the variances of the random effects are small relative to the residual variance. In this paper, we apply the ‘working parameter’ approach to derive alternative EM-type implementations for fitting mixed effects models, which we show empirically can be hundreds of times faster than the common EM-type implementations. In our limited simulations, they also compare well with the routines in S-PLUS® and Stata® in terms of both speed and reliability. The central idea of the working parameter approach is to search for efficient data augmentation schemes for implementing the EM algorithm by minimizing the augmented information over the working parameter, and in the mixed effects setting this leads to a transfer of the mixed effects variances into the regression slope parameters. We also describe a variation for computing the restricted maximum likelihood estimate and an adaptive algorithm that takes advantage of both the standard and the alternative EM-type implementations.

Some algebra and geometry for hierarchical models, applied to diagnostics

Abstract: Recent advances in computing make it practical to use complex hierarchical models. However, the complexity makes it difficult to see how features of the data determine the fitted model. This paper describes an approach to diagnostics for hierarchical models, specifically linear hierarchical models with additive normal or t-errors. The key is to express hierarchical models in the form of ordinary linear models by adding artificial `cases' to the data set corresponding to the higher levels of the hierarchy. The error term of this linear model is not homoscedastic, but its covariance structure is much simpler than that usually used in variance component or random effects models. The re-expression has several advantages. First, it is extremely general, covering dynamic linear models, random effect and mixed effect models, and pairwise difference models, among others. Second, it makes more explicit the geometry of hierarchical models, by analogy with the geometry of linear models. Third, the analogy with linear models provides a rich source of ideas for diagnostics for all the parts of hierarchical models. This paper gives diagnostics to examine candidate added variables, transformations, collinearity, case influence and residuals.

A stochastic approximation algorithm with Markov chain Monte-Carlo method for incomplete data estimation problems

Abstract: We propose a general procedure for solving incomplete data estimation problems. The procedure can be used to find the maximum likelihood estimate or to solve estimating equations in difficult cases such as estimation with the censored or truncated regression model, the nonlinear structural measurement error model, and the random effects model. The procedure is based on the general principle of stochastic approximation and the Markov chain Monte-Carlo method. Applying the theory on adaptive algorithms, we derive conditions under which the proposed procedure converges. Simulation studies also indicate that the proposed procedure consistently converges to the maximum likelihood estimate for the structural measurement error logistic regression model.