Estimation and prediction for low degree polynomial models under measurement errors with an application to forest harvesters

Summary.  In a modern computer-based forest harvester, tree stems are run in sequence through the measuring equipment root end first, and simultaneously the length and diameter are stored in a computer. These measurements may be utilized for example in the determination of the optimal cutting points of the stems. However, a problem that is often passed over is that these variables are usually measured with error. We consider estimation and prediction of stem curves when the length and diameter measurements are subject to errors. It is shown that only in the simplest case of a first-order model can the estimation be carried out unbiasedly by using standard least squares procedures. However, both the first- and the second-degree models are unbiased in prediction. Also a study on real stem is used to illustrate the models that are discussed.     

Large sample estimation and prediction for explosive growth curve models

Asymptotic distributions of maximum likelihood estimators for the parameters in explosive growth curve models are derived. Limit distributions of prediction errors when the parameters are estimated are also obtained. The growth curve models are viewed as multivariate time-series models, and the usual time-series methods are used for prediction. Estimation constrained by a hypothesis of homogeneity of growth rates is also considered.     

Large-sample comparisons of calibration procedures when both measurements are subject to error: the unreplicated case

A predictive functional relationship model is presented for the calibration problem in which the standard as well as the nonstandard measurements are subject to error. For the estimation of the relationship between the two measurements, the ordinary least squares and maximum likelihood estimation methods are considered, while for the prediction of unknown standard measurements we consider direct and inverse approaches. Relative performances of those calibration procedures are compared in terms of the asymptotic mean square error of prediction.     

Using relative utility curves to evaluate risk prediction

Summary.  Because many medical decisions are based on risk prediction models that are constructed from medical history and results of tests, the evaluation of these prediction models is important. This paper makes five contributions to this evaluation: the relative utility curve which gauges the potential for better prediction in terms of utilities, without the need for a reference level for one utility, while providing a sensitivity analysis for misspecification of utilities, the relevant region, which is the set of values of prediction performance that are consistent with the recommended treatment status in the absence of prediction, the test threshold, which is the minimum number of tests that would be traded for a true positive prediction in order for the expected utility to be non-negative, the evaluation of two-stage predictions that reduce test costs and connections between various measures of performance of prediction. An application involving the risk of cardiovascular disease is discussed.     

On modelling data from degradation sample paths over time

This paper is mainly concerned with modelling data from degradation sample paths over time. It uses a general growth curve model with Box‐Cox transformation, random effects and ARMA(p, q) dependence to analyse a set of such data. A maximum likelihood estimation procedure for the proposed model is derived and future values are predicted, based on the best linear unbiased prediction. The paper compares the proposed model with a nonlinear degradation model from a prediction point of view. Forecasts of failure times with various data lengths in the sample are also compared.     

Assessing discrimination of risk prediction rules in a clustered data setting

The AUC (area under ROC curve) is a commonly used metric to assess discrimination of risk prediction rules; however, standard errors of AUC are usually based on the Mann–Whitney U test that assumes independence of sampling units. For ophthalmologic applications, it is desirable to assess risk prediction rules based on eye-specific outcome variables which are generally highly, but not perfectly correlated in fellow eyes [e.g. progression of individual eyes to age-related macular degeneration (AMD)]. In this article, we use the extended Mann–Whitney U test (Rosner and Glynn, Biometrics 65:188–197, 2009) for the case where subunits within a cluster may have different progression status and assess discrimination of different prediction rules in this setting. Both data analyses based on progression of AMD and simulation studies show reasonable accuracy of this extended Mann–Whitney U test to assess discrimination of eye-specific risk prediction rules.     

A non-homogeneous hidden Markov model for predicting the distribution of sea surface elevation

The prediction problem of sea state based on the field measurements of wave and meteorological factors is a topic of interest from the standpoints of navigation safety and fisheries. Various statistical methods have been considered for the prediction of the distribution of sea surface elevation. However, prediction of sea state in the transitional situation when waves are developing by blowing wind has been a difficult problem until now, because the statistical expression of the dynamic mechanism during this situation is very complicated. In this article, we consider this problem through the development of a statistical model. More precisely, we develop a model for the prediction of the time-varying distribution of sea surface elevation, taking into account a non-homogeneous hidden Markov model in which the time-varying structures are influenced by wind speed and wind direction. Our prediction experiments suggest the possibility that the proposed model contributes to an improvement of the prediction accuracy by using a homogenous hidden Markov model. Furthermore, we found that the prediction accuracy is influenced by the circular distribution of the circular hidden Markov model for the directional time series wind direction data.     

On the relative efficiency of a monotone parameter curve estimator in a functional nonlinear model

Functional regression models that relate functional covariates to a scalar response are becoming more common due to the availability of functional data and computational advances. We introduce a functional nonlinear model with a scalar response where the true parameter curve is monotone. Using the Newton-Raphson method within a backfitting procedure, we discuss a penalized least squares criterion for fitting the functional nonlinear model with the smoothing parameter selected using generalized cross validation. Connections between a nonlinear mixed effects model and our functional nonlinear model are discussed, thereby providing an additional model fitting procedure using restricted maximum likelihood for smoothing parameter selection. Simulated relative efficiency gains provided by a monotone parameter curve estimator relative to an unconstrained parameter curve estimator are presented. In addition, we provide an application of our model with data from ozonesonde measurements of stratospheric ozone in which the measurements are biased as a function of altitude.     

Multilevel multivariate modelling of childhood growth, numbers of growth measurements and adult characteristics

Summary.  A general latent normal model for multilevel data with mixtures of response types is extended in the case of ordered responses to deal with variates having a large number of categories and including count data. An example is analysed by using repeated measures data on child growth and adult measures of body mass index and glucose. Applications are described that are concerned with the flexible prediction of adult measurements from collections of growth measurements and for studying the relationship between the number of measurement occasions and growth trajectories.     

Two transformation models for estimating an ROC curve derived from continuous data

A receiver operating characteristic (ROC) curve is a plot of two survival functions, derived separately from the diseased and healthy samples. A special feature is that the ROC curve is invariant to any monotone transformation of the measurement scale. We propose and analyse semiparametric and parametric transformation models for this two-sample problem. Following an unspecified or specified monotone transformation, we assume that the healthy and diseased measurements have two normal distributions with different means and variances. Maximum likelihood algorithms for estimating ROC curve parameters are developed. The proposed methods are illustrated on the marker CA125 in the diagnosis of gastric cancer.     

Bayesian analysis of a growth curve model with power transformation, random effects and AR(1) dependence

In this paper we devote ourselves to a general growth curve model with power transformation, random effects and AR(1) dependence via a Bayesian approach. Two priors are proposed and both parameter estimation and prediction of future values are considered. Some numerical results with a set of real data are also given.     

Assessing inter- and intra-agreement for dependent binary data: a Bayesian hierarchical correlation approach

Agreement measures are designed to assess consistency between different instruments rating measurements of interest. When the individual responses are correlated with multilevel structure of nestings and clusters, traditional approaches are not readily available to estimate the inter- and intra-agreement for such complex multilevel settings. Our research stems from conformity evaluation between optometric devices with measurements on both eyes, equality tests of agreement in high myopic status between monozygous twins and dizygous twins, and assessment of reliability for different pathologists in dysplasia. In this paper, we focus on applying a Bayesian hierarchical correlation model incorporating adjustment for explanatory variables and nesting correlation structures to assess the inter- and intra-agreement through correlations of random effects for various sources. This Bayesian generalized linear mixed-effects model (GLMM) is further compared with the approximate intra-class correlation coefficients and kappa measures by the traditional Cohen’s kappa statistic and the generalized estimating equations (GEE) approach. The results of comparison studies reveal that the Bayesian GLMM provides a reliable and stable procedure in estimating inter- and intra-agreement simultaneously after adjusting for covariates and correlation structures, in marked contrast to Cohen’s kappa and the GEE approach.     

Two-stage informative cluster sampling—estimation and prediction with applications for small-area models

This paper considers the effects of informative two-stage cluster sampling on estimation and prediction. The aims of this article are twofold: first to estimate the parameters of the superpopulation model for two-stage cluster sampling from a finite population, when the sampling design for both stages is informative, using maximum likelihood estimation methods based on the sample-likelihood function; secondly to predict the finite population total and to predict the cluster-specific effects and the cluster totals for clusters in the sample and for clusters not in the sample. To achieve this we derive the sample and sample-complement distributions and the moments of the first and second stage measurements. Also we derive the conditional sample and conditional sample-complement distributions and the moments of the cluster-specific effects given the cluster measurements. It should be noted that classical design-based inference that consists of weighting the sample observations by the inverse of sample selection probabilities cannot be applied for the prediction of the cluster-specific effects for clusters not in the sample. Also we give an alternative justification of the Royall [1976. The linear least squares prediction approach to two-stage sampling. Journal of the American Statistical Association 71, 657–664] predictor of the finite population total under two-stage cluster population. Furthermore, small-area models are studied under informative sampling.     

Predicting objective physical activity from self-report surveys: a model validation study using estimated generalized least-squares regression

Physical activity measurements derived from self-report surveys are prone to measurement errors. Monitoring devices like accelerometers offer more objective measurements of physical activity, but are impractical for use in large-scale surveys. A model capable of predicting objective measurements of physical activity from self-reports would offer a practical alternative to obtaining measurements directly from monitoring devices. Using data from National Health and Nutrition Examination Survey 2003–2006, we developed and validated models for predicting objective physical activity from self-report variables and other demographic characteristics. The prediction intervals produced by the models were large, suggesting that the ability to predict objective physical activity for individuals from self-reports is limited.     

Aggregated functional data model for near-infrared spectroscopy calibration and prediction

Calibration and prediction for NIR spectroscopy data are performed based on a functional interpretation of the Beer–Lambert formula. Considering that, for each chemical sample, the resulting spectrum is a continuous curve obtained as the summation of overlapped absorption spectra from each analyte plus a Gaussian error, we assume that each individual spectrum can be expanded as a linear combination of B-splines basis. Calibration is then performed using two procedures for estimating the individual analytes’ curves: basis smoothing and smoothing splines. Prediction is done by minimizing the square error of prediction. To assess the variance of the predicted values, we use a leave-one-out jackknife technique. Departures from the standard error models are discussed through a simulation study, in particular, how correlated errors impact on the calibration step and consequently on the analytes’ concentration prediction. Finally, the performance of our methodology is demonstrated through the analysis of two publicly available datasets.     

A test for crossing receiver operating characteristic (roc) curves

The receiver operating characteristic (ROC) curve gives a graphical representation of sensitivity and specificity of a prediction model when varying the decision treshold on a diagnostic criterion. A classical test for comparing the overall accuracies for two models -1 and 2- is based on the difference between ROC curves areas - related to its standard error. This test is designed for the situation where ROC curve 1 caps ROC curve 2. Often both curves cross :in this paper, a new test, based on the integrated difference between the curves, is proposed to deal with this situation. In a simulation experiment, the new test was less powerful than the old test for detecting an overall superiority, but much more powerfull against the crossing alternative.     

Curve prediction and clustering with mixtures of Gaussian process functional regression models

Shi, Wang, Murray-Smith and Titterington (Biometrics 63:714–723, 2007) proposed a Gaussian process functional regression (GPFR) model to model functional response curves with a set of functional covariates. Two main problems are addressed by their method: modelling nonlinear and nonparametric regression relationship and modelling covariance structure and mean structure simultaneously. The method gives very good results for curve fitting and prediction but side-steps the problem of heterogeneity. In this paper we present a new method for modelling functional data with ‘spatially’ indexed data, i.e., the heterogeneity is dependent on factors such as region and individual patient’s information. For data collected from different sources, we assume that the data corresponding to each curve (or batch) follows a Gaussian process functional regression model as a lower-level model, and introduce an allocation model for the latent indicator variables as a higher-level model. This higher-level model is dependent on the information related to each batch. This method takes advantage of both GPFR and mixture models and therefore improves the accuracy of predictions. The mixture model has also been used for curve clustering, but focusing on the problem of clustering functional relationships between response curve and covariates, i.e. the clustering is based on the surface shape of the functional response against the set of functional covariates. The model is examined on simulated data and real data.     

A Basis Function Approach to Position Estimation Using Microwave Arrays

We consider the problem of estimating the bearing of a remote object given measurements on a particular type of non-scanning radar, namely a focal-plane array. Such a system focuses incoming radiation through a lens onto an array of detectors. The problem is to estimate the angular position of the radiation source given measurements on the array of detectors and knowledge of the properties of the lens. The training data are essentially noiseless, and an estimator is derived for noisy test conditions. An approach based on kernel basis functions is developed. The estimate of the basis function weights is achieved through a regularization or roughness penalty approach. Choosing the regularization parameter to be proportional to the inverse of the input signal-to-noise ratio leads to a minimum prediction error. Experimental results for a 12-element detector array support the theoretical predictions.     

Borrowing Information across Populations in Estimating Positive and Negative Predictive Values

A marker's capacity to predict risk of a disease depends on disease prevalence in the target population and its classification accuracy, i.e. its ability to discriminate diseased subjects from non-diseased subjects. The latter is often considered an intrinsic property of the marker; it is independent of disease prevalence and hence more likely to be similar across populations than risk prediction measures. In this paper, we are interested in evaluating the population-specific performance of a risk prediction marker in terms of positive predictive value (PPV) and negative predictive value (NPV) at given thresholds, when samples are available from the target population as well as from another population. A default strategy is to estimate PPV and NPV using samples from the target population only. However, when the marker's classification accuracy as characterized by a specific point on the receiver operating characteristics (ROC) curve is similar across populations, borrowing information across populations allows increased efficiency in estimating PPV and NPV. We develop estimators that optimally combine information across populations. We apply this methodology to a cross-sectional study where we evaluate PCA3 as a risk prediction marker for prostate cancer among subjects with or without previous negative biopsy.     

Propensity score prediction for electronic healthcare databases using super learner and high-dimensional propensity score methods

ABSTRACT

The optimal learner for prediction modeling varies depending on the underlying data-generating distribution. Super Learner (SL) is a generic ensemble learning algorithm that uses cross-validation to select among a ‘library’ of candidate prediction models. While SL has been widely studied in a number of settings, it has not been thoroughly evaluated in large electronic healthcare databases that are common in pharmacoepidemiology and comparative effectiveness research. In this study, we applied and evaluated the performance of SL in its ability to predict the propensity score (PS), the conditional probability of treatment assignment given baseline covariates, using three electronic healthcare databases. We considered a library of algorithms that consisted of both nonparametric and parametric models. We also proposed a novel strategy for prediction modeling that combines SL with the high-dimensional propensity score (hdPS) variable selection algorithm. Predictive performance was assessed using three metrics: the negative log-likelihood, area under the curve (AUC), and time complexity. Results showed that the best individual algorithm, in terms of predictive performance, varied across datasets. The SL was able to adapt to the given dataset and optimize predictive performance relative to any individual learner. Combining the SL with the hdPS was the most consistent prediction method and may be promising for PS estimation and prediction modeling in electronic healthcare databases.