Home About us Contact
|
Home About us Contact | ||
|
|
||
Conditional Expectation (conditional + expectation)
Selected AbstractsModeling the Effects of a Service Guarantee on Perceived Service Quality Using Alternating Conditional Expectations (ACE),DECISION SCIENCES, Issue 3 2002Chee-Chuong Sum ABSTRACT This paper addresses the dearth of empirical research on the relationship between service guarantee and perceived service quality (PSQ). In particular, we examine the moderating effects of a service guarantee on PSQ. While a recent study provided empirical evidence that service quality is affected by service guarantee and employee variables such as employee motivation/vision and learning through service failure, the nature and form of the relationships between these variables remain unclear. Knowledge of these relationships can assist service managers to allocate resources more judiciously, avoid pitfalls, and establish more realistic expectations. Data was obtained from employees and customers of a multinational hotel chain that has implemented a service guarantee program in 89 of its hotels in America and Canada. As the employee variables could affect performance in a non-linear fashion, we relaxed the assumption of model linearity by using the Alternating Conditional Expectations (ACE) algorithm to arrive at a better-fitting, non-linear regression model for PSQ. Our findings indicate the existence of significant non-linear relationships between PSQ and its determinant variables. The ACE model also revealed that service guarantee interacts with the employee variables to affect PSQ in a non-linear fashion. The non-linear relationships present new insights into the management of service guarantees and PSQ. Explanations and managerial implications of our results are presented and discussed. [source] OUTLYING OBSERVATIONS AND MISSING VALUES: HOW SHOULD THEY BE HANDLED?CLINICAL AND EXPERIMENTAL PHARMACOLOGY AND PHYSIOLOGY, Issue 5-6 2008John Ludbrook SUMMARY 1The problems of, and best solutions for, outlying observations and missing values are very dependent on the sizes of the experimental groups. For original articles published in Clinical and Experimental Pharmacology and Physiology during 2006,2007, the range of group sizes ranged from three to 44 (,small groups'). In surveys, epidemiological studies and clinical trials, the group sizes range from 100s to 1000s (,large groups'). 2How can one detect outlying (extreme) observations? The best methods are graphical, for instance: (i) a scatterplot, often with mean±2 s; and (ii) a box-and-whisker plot. Even with these, it is a matter of judgement whether observations are truly outlying. 3It is permissable to delete or replace outlying observations if an independent explanation for them can be found. This may be, for instance, failure of a piece of measuring equipment or human error in operating it. If the observation is deleted, it can then be treated as a missing value. Rarely, the appropriate portion of the study can be repeated. 4It is decidedly not permissable to delete unexplained extreme values. Some of the acceptable strategies for handling them are: (i) transform the data and proceed with conventional statistical analyses; (ii) use the mean for location, but use permutation (randomization) tests for comparing means; and (iii) use robust methods for describing location (e.g. median, geometric mean, trimmed mean), for indicating dispersion (range, percentiles), for comparing locations and for regression analysis. 5What can be done about missing values? Some strategies are: (i) ignore them; (ii) replace them by hand if the data set is small; and (iii) use computerized imputation techniques to replace them if the data set is large (e.g. regression or EM (conditional Expectation, Maximum likelihood estimation) methods). 6If the missing values are ignored, or even if they are replaced, it is essential to test whether the individuals with missing values are otherwise indistinguishable from the remainder of the group. If the missing values have not occurred at random, but are associated with some property of the individuals being studied, the subsequent analysis may be biased. [source] PAIRWISE DIFFERENCE ESTIMATION WITH NONPARAMETRIC CONTROL VARIABLES,INTERNATIONAL ECONOMIC REVIEW, Issue 4 2007Andres Aradillas-Lopez This article extends the pairwise difference estimators for various semilinear limited dependent variable models proposed by Honoré and Powell (Identification and Inference in Econometric Models. Essays in Honor of Thomas Rothenberg Cambridge: Cambridge University Press, 2005) to permit the regressor appearing in the nonparametric component to itself depend upon a conditional expectation that is nonparametrically estimated. This permits the estimation approach to be applied to nonlinear models with sample selectivity and/or endogeneity, in which a "control variable" for selectivity or endogeneity is nonparametrically estimated. We develop the relevant asymptotic theory for the proposed estimators and we illustrate the theory to derive the asymptotic distribution of the estimator for the partially linear logit model. [source] On the Construction of Imputation Classes in SurveysINTERNATIONAL STATISTICAL REVIEW, Issue 1 2007David Haziza Summary This paper explores the problem of the construction of imputation classes using the score method, sometimes called predictive mean stratification or response propensity stratification, depending on the context. This method was studied in Thomsen (1973), Little (1986) and Eltinge & Yansaneh (1997). We use a different framework to evaluate the properties of the resulting imputed estimator of a population mean. In our framework, we condition on the realized sample. This enables us to considerably simplify our theoretical developments in the frequent situation where the boundaries and the number of classes are sample-dependent. We find that the key factor for reducing the non-response bias is to form classes homogeneous with respect to the response probabilities and/or the conditional expectation of the variable of interest. In the latter case, the non-response/imputation variance is also reduced. Finally, we performed a simulation study to fully evaluate various versions of the score method and to compare them with a cross-classification method, which is frequently used in practice. The results showed the superiority of the score method in general. Résumé Cet article étudie la construction des classes d'imputation par la méthode des scores, appelée également stratification par moyenne prédite ou stratification par propensité de réponse selon le contexte. Cette méthode a étéétudiée par Thomsen (1973), Little (1986) et Eltinge et Yansaneh (1997). Nous utilisons un cadre de travail différent permettant d'évaluer les propriétés de l'estimateur imputé de la moyenne de la population selon lequel nous conditionnons sur l'échantillon réalisé. Ceci nous permet de simplifier considérablement les développements théoriques lorsque les bornes et le nombre de classes dépendent de l'échantillon, ce qui survient fréquemment en pratique. Nous déterminons que le facteur clé permettant de réduire le biais du à la non-réponse est de former des classes qui soient homogènes par rapport aux probabilités de réponse et/ou à l'espérance conditionnelle de la variable d'intérêt. Dans ce dernier cas, la variance due à la non-réponse et à l'imputation est également réduite. Finalement, nous effectuons une étude par simulation afin d'évaluer en profondeur plusieurs versions de la méthode des scores et de comparer celles-ci avec la méthode par croisement qui est fréquemment utilisée en pratique. Les résultats obtenus montrent la supériorité de la méthode des scores en général. [source] A Partially Observed Model for Micromovement of Asset Prices with Bayes Estimation via FilteringMATHEMATICAL FINANCE, Issue 3 2003Yong Zeng A general micromovement model that describes transactional price behavior is proposed. The model ties the sample characteristics of micromovement and macromovement in a consistent manner. An important feature of the model is that it can be transformed to a filtering problem with counting process observations. Consequently, the complete information of price and trading time is captured and then utilized in Bayes estimation via filtering for the parameters. The filtering equations are derived. A theorem on the convergence of conditional expectation of the model is proved. A consistent recursive algorithm is constructed via the Markov chain approximation method to compute the approximate posterior and then the Bayes estimates. A simplified model and its recursive algorithm are presented in detail. Simulations show that the computed Bayes estimates converge to their true values. The algorithm is applied to one month of intraday transaction prices for Microsoft and the Bayes estimates are obtained. [source] Measuring Conditional Persistence in Nonlinear Time Series,OXFORD BULLETIN OF ECONOMICS & STATISTICS, Issue 3 2007George Kapetanios Abstract The persistence properties of economic time series have been a primary object of investigation in a variety of guises since the early days of econometrics. Recently, work on nonlinear modelling for time series has introduced the idea that persistence of a shock at a point in time may vary depending on the state of the process at that point in time. This article suggests investigating the persistence of processes conditioning on their history as a tool that may aid parametric nonlinear modelling. In particular, we suggest that examining the nonparametrically estimated derivatives of the conditional expectation of a variable with respect to its lag(s) may be a useful indicator of the variation in persistence with respect to its past history. We discuss in detail the implementation of the measure and present a Monte Carlo investigation. We further apply the persistence analysis to real exchange rates. [source] Robustified Maximum Likelihood Estimation in Generalized Partial Linear Mixed Model for Longitudinal DataBIOMETRICS, Issue 1 2009Guo You Qin Summary In this article, we study the robust estimation of both mean and variance components in generalized partial linear mixed models based on the construction of robustified likelihood function. Under some regularity conditions, the asymptotic properties of the proposed robust estimators are shown. Some simulations are carried out to investigate the performance of the proposed robust estimators. Just as expected, the proposed robust estimators perform better than those resulting from robust estimating equations involving conditional expectation like Sinha (2004, Journal of the American Statistical Association99, 451,460) and Qin and Zhu (2007, Journal of Multivariate Analysis98, 1658,1683). In the end, the proposed robust method is illustrated by the analysis of a real data set. [source] Identification of neutrophil gelatinase-associated lipocalin (NGAL) as a discriminatory marker of the hepatocyte-secreted protein response to IL-1,: a proteomic analysisBIOTECHNOLOGY & BIOENGINEERING, Issue 4 2005Arul Jayaraman Abstract The liver is the major source of proteins used throughout the body for various functions. Upon injury or infection, an acute phase response (APR) is initiated in the liver that is primarily mediated by inflammatory cytokines such as interleukin-1, (IL-1,) and interleukin-6. Among others, the APR is characterized by an altered protein synthetic profile. We used two-dimensional gel electrophoresis to study the dynamics of changes in protein synthesis in hepatocytes exposed to these inflammatory cytokines. Protein profiles were quantified using image analysis and further analyzed using multivariate statistical methods. Our results indicate that IL-1, and IL-6 each induces secreted protein responses with distinct dynamics and dose-dependence. Parallel stimulation by IL-1, and IL-6 results in a protein pattern indistinguishable from the IL-1, pattern, indicating a dominant effect of IL-1, over IL-6 at the doses tested. Multidimensional scaling (MDS) of correlation distances between protein secretion levels revealed two protein pairs that are robustly co-secreted across the various cytokine stimulation conditions, suggesting shared regulatory pathways. Finally, we also used multivariate alternating conditional expectation (MACE) to identify transformation functions that discriminated the cytokine-stimulated and untreated hepatocyte-secreted protein profiles. Our analysis indicates that the expression of neutrophil gelatinase-associated lipocalin (NGAL) was sufficient to discriminate between IL-1, and IL-6 stimulation. The combination of proteomics and multivariate analysis is expected to provide new information on the cellular regulatory networks involved in generating specific cellular responses. © 2005 Wiley Periodicals, Inc. [source] Maximum likelihood estimation of higher-order integer-valued autoregressive processesJOURNAL OF TIME SERIES ANALYSIS, Issue 6 2008Ruijun Bu Abstract., In this article, we extend the earlier work of Freeland and McCabe [Journal of time Series Analysis (2004) Vol. 25, pp. 701,722] and develop a general framework for maximum likelihood (ML) analysis of higher-order integer-valued autoregressive processes. Our exposition includes the case where the innovation sequence has a Poisson distribution and the thinning is binomial. A recursive representation of the transition probability of the model is proposed. Based on this transition probability, we derive expressions for the score function and the Fisher information matrix, which form the basis for ML estimation and inference. Similar to the results in Freeland and McCabe (2004), we show that the score function and the Fisher information matrix can be neatly represented as conditional expectations. Using the INAR(2) specification with binomial thinning and Poisson innovations, we examine both the asymptotic efficiency and finite sample properties of the ML estimator in relation to the widely used conditional least squares (CLS) and Yule,Walker (YW) estimators. We conclude that, if the Poisson assumption can be justified, there are substantial gains to be had from using ML especially when the thinning parameters are large. [source] First-Order Schemes in the Numerical Quantization MethodMATHEMATICAL FINANCE, Issue 1 2003V. Bally The numerical quantization method is a grid method that relies on the approximation of the solution to a nonlinear problem by piecewise constant functions. Its purpose is to compute a large number of conditional expectations along the path of the associated diffusion process. We give here an improvement of this method by describing a first-order scheme based on piecewise linear approximations. Main ingredients are correction terms in the transition probability weights. We emphasize the fact that in the case of optimal quantization, many of these correcting terms vanish. We think that this is a strong argument to use it. The problem of pricing and hedging American options is investigated and a priori estimates of the errors are proposed. [source] Optimal Malliavin Weighting Function for the Computation of the GreeksMATHEMATICAL FINANCE, Issue 1 2003Eric Benhamou This paper reexamines the Malliavin weighting functions introduced by Fournié et al. (1999) as a new method for efficient and fast computations of the Greeks. Reexpressing the weighting function generator in terms of its Skorohod integrand, we show that these weighting functions have to satisfy necessary and sufficient conditions expressed as conditional expectations. We then derive the weighting function with the smallest total variance. This is of particular interest as it bridges the method of Malliavin weights and the one of likelihood ratio, as introduced by Broadie and Glasserman (1996). The likelihood ratio is precisely the weighting function with the smallest total variance. We finally examine when to use the Malliavin method and when to prefer finite difference. [source] A Version of the EM Algorithm for Proportional Hazard Model with Random EffectsBIOMETRICAL JOURNAL, Issue 6 2005José Cortiñas Abrahantes Abstract Proportional hazard models with multivariate random effects (frailties) acting multiplicatively on the baseline hazard have recently become a topic of an intensive research. One of the main practical problems related to the models is the estimation of parameters. To this aim, several approaches based on the EM algorithm have been proposed. The major difference between these approaches is the method of the computation of conditional expectations required at the E-step. In this paper an alternative implementation of the EM algorithm is proposed, in which the expected values are computed with the use of the Laplace approximation. The method is computationally less demanding than the approaches developed previously. Its performance is assessed based on a simulation study and compared to a non-EM based estimation approach proposed by Ripatti and Palmgren (2000). (© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim) [source] | |||||||||||||