Home About us Contact
|
Home About us Contact | ||
|
|
||
Repeated Observations (repeated + observation)
Selected AbstractsTechnical Note: Grading the vertical cup:disc ratio and the effect of scalingOPHTHALMIC AND PHYSIOLOGICAL OPTICS, Issue 6 2007Ruth Bennett Abstract Purpose:, To evaluate the effect of scaling on sensitivity to change for grading the vertical cup:disc ratio (CDR). Methods:, Vertical CDR was assessed by six observers (three ophthalmologists and three optometrists) on 43 stereo disc photographs. Repeated observations were made for both 0.1 and 0.05 interval scales. Paired differences were calculated for all observers and each observer separately. Mean and standard deviation of differences and agreement statistics were used to compare scales. Results:, Five observers demonstrated a reduction in the spread of differences (mean difference 0.19 to 0.15) and all observers demonstrated a reduction in concordance using the finer scale (mean concordance 54% to 39%). Conclusion:, The use of a finer scale reduces test,retest variability and increases sensitivity to change when estimating the vertical CDR. Use of this scale does not require any additional resource and it may be easily implemented in routine clinical practice. [source] Estimation of Nonlinear Models with Measurement ErrorECONOMETRICA, Issue 1 2004Susanne M. Schennach This paper presents a solution to an important econometric problem, namely the root n consistent estimation of nonlinear models with measurement errors in the explanatory variables, when one repeated observation of each mismeasured regressor is available. While a root n consistent estimator has been derived for polynomial specifications (see Hausman, Ichimura, Newey, and Powell (1991)), such an estimator for general nonlinear specifications has so far not been available. Using the additional information provided by the repeated observation, the suggested estimator separates the measurement error from the "true" value of the regressors thanks to a useful property of the Fourier transform: The Fourier transform converts the integral equations that relate the distribution of the unobserved "true" variables to the observed variables measured with error into algebraic equations. The solution to these equations yields enough information to identify arbitrary moments of the "true," unobserved variables. The value of these moments can then be used to construct any estimator that can be written in terms of moments, including traditional linear and nonlinear least squares estimators, or general extremum estimators. The proposed estimator is shown to admit a representation in terms of an influence function, thus establishing its root n consistency and asymptotic normality. Monte Carlo evidence and an application to Engel curve estimation illustrate the usefulness of this new approach. [source] Longitudinal data analysis in pedigree studiesGENETIC EPIDEMIOLOGY, Issue S1 2003W. James Gauderman Abstract Longitudinal family studies provide a valuable resource for investigating genetic and environmental factors that influence long-term averages and changes over time in a complex trait. This paper summarizes 13 contributions to Genetic Analysis Workshop 13, which include a wide range of methods for genetic analysis of longitudinal data in families. The methods can be grouped into two basic approaches: 1) two-step modeling, in which repeated observations are first reduced to one summary statistic per subject (e.g., a mean or slope), after which this statistic is used in a standard genetic analysis, or 2) joint modeling, in which genetic and longitudinal model parameters are estimated simultaneously in a single analysis. In applications to Framingham Heart Study data, contributors collectively reported evidence for genes that affected trait mean on chromosomes 1, 2, 3, 5, 8, 9, 10, 13, and 17, but most did not find genes affecting slope. Applications to simulated data suggested that even for a gene that only affected slope, use of a mean-type statistic could provide greater power than a slope-type statistic for detecting that gene. We report on the results of a small experiment that sheds some light on this apparently paradoxical finding, and indicate how one might form a more powerful test for finding a slope-affecting gene. Several areas for future research are discussed. Genet Epidemiol 25 (Suppl. 1):S18,S28, 2003. © 2003 Wiley-Liss, Inc. [source] Social influence on predictions of simulated stock pricesJOURNAL OF BEHAVIORAL DECISION MAKING, Issue 3 2009Maria Andersson Abstract Herding in financial markets refers to that investors are influenced by others. This study addresses the importance of consistency for herding. It is suggested that, in financial markets perceptions of consistency are based on repeated observations over time. Consistency may then be perceived as the agreement across time between investors' predictions. In addition, consistency may be related to variance over time in each investor's predictions. In an experiment using a Multiple Cue Probability Learning paradigm, 96 undergraduates made multi-trial predictions of future stock prices given information about the current price and the predictions made by five fictitious others. Consistency was varied between the others' predictions (correlation) and within the others' predictions (variance). The results showed that the predictions were significantly influenced by the others' predictions when these were correlated. No effect of variance was observed. Hence, participants were influenced by the others when they were in agreement, regardless of whether they varied their predictions over trials or not. Copyright © 2008 John Wiley & Sons, Ltd. [source] Bonus-Malus Scales in Segmented Tariffs With Stochastic Migration Between SegmentsJOURNAL OF RISK AND INSURANCE, Issue 4 2003Natacha Brouhns This article proposes a computer-intensive methodology to build bonus-malus scales in automobile insurance. The claim frequency model is taken from Pinquet, Guillén, and Bolancé (2001). It accounts for overdispersion, heteroskedasticity, and dependence among repeated observations. Explanatory variables are taken into account in the determination of the relativities, yielding an integrated automobile ratemaking scheme. In that respect, it complements the study of Taylor (1997). [source] A hierarchical Bayesian model for predicting the functional consequences of amino-acid polymorphismsJOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES C (APPLIED STATISTICS), Issue 1 2005Claudio J. Verzilli Summary., Genetic polymorphisms in deoxyribonucleic acid coding regions may have a phenotypic effect on the carrier, e.g. by influencing susceptibility to disease. Detection of deleterious mutations via association studies is hampered by the large number of candidate sites; therefore methods are needed to narrow down the search to the most promising sites. For this, a possible approach is to use structural and sequence-based information of the encoded protein to predict whether a mutation at a particular site is likely to disrupt the functionality of the protein itself. We propose a hierarchical Bayesian multivariate adaptive regression spline (BMARS) model for supervised learning in this context and assess its predictive performance by using data from mutagenesis experiments on lac repressor and lysozyme proteins. In these experiments, about 12 amino-acid substitutions were performed at each native amino-acid position and the effect on protein functionality was assessed. The training data thus consist of repeated observations at each position, which the hierarchical framework is needed to account for. The model is trained on the lac repressor data and tested on the lysozyme mutations and vice versa. In particular, we show that the hierarchical BMARS model, by allowing for the clustered nature of the data, yields lower out-of-sample misclassification rates compared with both a BMARS and a frequen-tist MARS model, a support vector machine classifier and an optimally pruned classification tree. [source] Prediction of formation temperatures in permafrost regions from temperature logs in deep wells,field casesPERMAFROST AND PERIGLACIAL PROCESSES, Issue 3 2003I. M. Kutasov Abstract Important data on the thermal regime of the Earth's interior come from temperature measurements in deep boreholes. Drilling greatly alters the temperature field of earth materials surrounding the wellbore. In permafrost regions, due to thawing of adjacent strata during drilling, representative data can be obtained only by repeated observations over a long period of time. In this paper we predict undisturbed formation temperatures (and geothermal gradients) from shut-in temperature logs in deep wells. The main features of the method are: (1) in the permafrost section of the well, the starting point in the well thermal recovery is moved from the end of well completion to the moment of time when the refreezing of enclosing strata was completed; it takes into account the refreezing of thawed material in a temperature interval; and (2) below the permafrost base, the starting point in the well thermal recovery is moved from the end of well completion to the moment of time when the first shut-in temperature log was taken. A generalized formula to process field data (for the well sections below and above the permafrost base) is presented. Temperature logs conducted in five wells verify the method. Copyright © 2003 John Wiley & Sons, Ltd. [source] Using Multimember District Elections to Estimate the Sources of the Incumbency AdvantageAMERICAN JOURNAL OF POLITICAL SCIENCE, Issue 2 2009Shigeo Hirano In this article we use a novel research design that exploits unique features of multimember districts to estimate and decompose the incumbency advantage in state legislative elections. Like some existing related studies we also use repeated observations on the same candidates to account for unobserved factors that remain constant across observations. Multimember districts have the additional feature of copartisans competing for multiple seats within the same district. This allows us to identify both the direct office-holder benefits and the incumbent quality advantage over nonincumbent candidates from the same party. We find that the overall incumbency advantage is of similar magnitude as that found in previous studies. We attribute approximately half of this advantage to incumbents' quality advantage over open-seat candidates and the remainder to direct office-holder benefits. However, we also find some evidence that direct office-holder benefits are larger in competitive districts than in safe districts and in states with relatively large legislative budgets per capita. [source] Omitted variables in longitudinal data modelsTHE CANADIAN JOURNAL OF STATISTICS, Issue 4 2001Edward W. Frees Abstract The omission of important variables is a well-known model specification issue in regression analysis and mixed linear models. The author considers longitudinal data models that are special cases of the mixed linear models; in particular, they are linear models of repeated observations on a subject. Models of omitted variables have origins in both the econometrics and biostatistics literatures. The author describes regression coefficient estimators that are robust to and that provide the basis for detecting the influence of certain types of omitted variables. New robust estimators and omitted variable tests are introduced and illustrated with a case study that investigates the determinants of tax liability. [source] Endothelin-1 gene polymorphism and hearing impairment in elderly JapaneseTHE LARYNGOSCOPE, Issue 5 2009Yasue Uchida MD Abstract Objectives/Hypothesis: To investigate the association between the Lys198Asn (G/T) polymorphism (rs5370) in the endothelin-1 gene (EDN1) and hearing impairment in middle-aged and elderly Japanese. Study Design: Longitudinal study. Methods: Data were collected from community-dwelling Japanese adults who participated in the Longitudinal Study of Aging biennially between 1997 and 2006. The participants at baseline were 2,231 adults aged 40 years to 79 years. An average hearing threshold level of 25 dB or better in the better ear for frequencies 500 Hz, 1,000 Hz, 2,000 Hz, and 4,000 Hz was defined as no hearing impairment. Using generalized estimating equations to treat repeated observations within subjects, 7,097 cumulative data were analyzed to assess the association between hearing status and the EDN1 G/T polymorphism with adjustment for age, sex, histories of ear disease, occupational noise exposure, heart disease, hypertension, and body mass index under additive, dominant, and recessive genetic models. Results: Comparison with wild-type homozygotes (GG), heterozygotes, and mutant homozygotes (GT/TT) showed a positive association with hearing impairment after adjustment for age in model 1 (odds ratio [OR] = 1.24; 95% confidence interval [CI] = 1.02,1.50; P = .033), for age and sex in model 2 (OR = 1.29; CI = 1.06,1.57; P = .0122), and for age, sex, history of ear disease, and history of occupational noise exposure in model 3 (OR = 1.31; CI = 1.07,1.60; P = .0092). The association was also significant in model 3 under the additive model. Conclusions: This study demonstrated that mutant T-allele carriers were associated with a higher risk of hearing impairment than carriers of wild-type homozygotes in middle-aged and elderly people. This result implies that endothelin-1 plays a valuable role in the cochlea. Laryngoscope, 2009 [source] Some contributions to the analysis of multivariate dataBIOMETRICAL JOURNAL, Issue 2 2009Arne C. Bathke Abstract In this paper, we provide an overview of recently developed methods for the analysis of multivariate data that do not necessarily emanate from a normal universe. Multivariate data occur naturally in the life sciences and in other research fields. When drawing inference, it is generally recommended to take the multivariate nature of the data into account, and not merely analyze each variable separately. Furthermore, it is often of major interest to select an appropriate set of important variables. We present contributions in three different, but closely related, research areas: first, a general approach to the comparison of mean vectors, which allows for profile analysis and tests of dimensionality; second, non-parametric and parametric methods for the comparison of independent samples of multivariate observations; and third, methods for the situation where the experimental units are observed repeatedly, for example, over time, and the main focus is on analyzing different time profiles when the number p of repeated observations per subject is larger than the number n of subjects. [source] | |||||||||||||||||||