Classical Estimation Approaches Research Articles

Survival Time Analysis of Hypertension Patients Using Parametric Models

Background: Hypertension is a worldwide public-health challenge and one of a leading modifiable risk factor for cardiovascular disease and death.
 Aims: The aim of this study was compare parameter estimations using both Bayesian and classical approaches and to detect out potential factors that affects survival probability of hypertension patient’s under follow up.
 Materials and Methods: A simple random sampling technique was used to select 430 patients among a total of 2126 hypertension patients who had been under follow up at Yekatit-12 Hospital in Ethiopia from January 2013 to January 2019. Parametric distributions: Exponential, Weibull, Lognormal and loglogistic are studied to analysis survival probabilities of the patients in both Bayesian and classical approaches. The model selection criteria are employed to identify the model with best fit to the data. Bayesian estimation approach was smaller deviance information criteria as compare to classical estimation approaches for the current data set.
 Results and Conclusion: The analysis Bayesian Weibull results indicate that the baseline age of the patient, gender, family history of hypertension, tobacco use, alcohol use, khat intake, blood cholesterol level of the patient, hypertension disease stage, adherence to the treatment and related disease were significantly associated with survival time of hypertension patients. Patients with raised blood Cholesterol level at baseline tend to have shorter survival time as compare to one with normal blood cholesterol level at baseline. Society and all stakeholders should be aware of the consequences of these factors which can influence the survival time of hypertension patients.

A Semiempirical Method to Estimate Actual Evapotranspiration in Mediterranean Environments

Actual evapotranspiration (ETA) is a major term of site water balance whose knowledge is essential for numerous purposes. The classical ETA estimation approach based on the use of multitemporal crop coefficients (Kc) cannot be applied in water-limited environments without proper correction. Such correction can be theoretically obtained by means of soil water content (SWC) measurements, which, however, are affected by several drawbacks, due to both their technical and operational characteristics. The current paper proposes a method to normalize annual SWC datasets and integrate them in an ETA estimation procedure suitable for monitoring both agricultural and natural Mediterranean ecosystems. Differently from previous approaches, the SWC normalization is obtained using data-specific information, which renders the new method mostly insensitive to the mentioned problems. The method is first described and then applied in three case studies representative of different Mediterranean ecosystems (i.e., grassland, coniferous, and deciduous forests). The results are evaluated versus latent heat measurements taken by eddy covariance flux towers. Satisfactory accuracy is obtained in all three case studies, with advantages and limitations which are discussed in the final concluding sections.

A beta-binomial mixed-effects model approach for analysing longitudinal discrete and bounded outcomes.

Patient-reported outcomes (PROs) are currently being increasingly used as primary outcome measures in observational and experimental studies since they inform clinicians and researchers about the health-status of patients and generate data to facilitate improved care. PROs usually appear as discrete and bounded with U, J, or inverse J shapes, and hence, exponential family members offer inadequate distributional fits. The beta-binomial distribution has been proposed in the literature to fit PROs. However, the fact that the beta-binomial distribution does not belong to the exponential family limits its applicability in the regression model context, and classical estimation approaches are not straightforward. Moreover, PROs are usually measured in a longitudinal framework in which individuals are followed up for a certain period. Hence, each individual obtains several scores of the PRO over time, which leads to the repeated measures and defines the correlation structure in the data. In this work, we have developed and proposed an estimation procedure for the analysis of correlated discrete and bounded outcomes, particularly PROs, by a beta-binomial mixed-effects model. Additionally, we have implemented the methodology in the PROreg package in R. Because there are similar approaches in the literature to address the same issue, this work also incorporates a comparison study between our proposal and alternative methodologies commonly implemented in R and shows the superior performance of our estimation procedure. This paper was motivated by the analysis of the health-status of patients with chronic obstructive pulmonary disease, where the main objective is the assessment of risk factors that may affect the evolution of the disease. The application of the proposed approach in the study leads to clinically relevant results.

Differences Between Classical and Bayesian Estimates for Mixed Logit Models: A Replication Study

SummaryThe mixed logit model is widely used in applied econometrics. Researchers typically rely on the free choice between the classical and Bayesian estimation approach. However, empirical evidence of the similarity of their parameter estimates is sparse. The presumed similarity is mainly based on one empirical study that analyzes a single dataset (Huber J, Train KE. 2001. On the similarity of classical and Bayesian estimates of individual mean partworths. Marketing Letters12(3): 259–269). Our replication study offers a generalization of their results by comparing classical and Bayesian parameter estimates from six additional datasets and specifically for panel versus cross‐sectional data. In general, our results suggest that the two methods provide similar results, with less similarity for cross‐sectional data than for panel data. Copyright © 2016 John Wiley & Sons, Ltd.

Using the photoload technique with double sampling to improve surface fuel loading estimates

Accurate surface fuel load estimates based on the planar intercept method require a considerable amount of time and cost. Recently the photoload method has been proposed as an alternative for sampling of fine woody surface fuels. To evaluate the use of photoload fuel sampling, six simulated fuel beds of 100 photoload visual estimates and destructively sampled fuel loads were generated at three levels of fuel loading (0.016, 0.060, and 0.120 kg m–2) and two levels of variability (coefficients of variation of ~42 and 85%). We assessed the accuracy and precision of simple random sampling with and without double sampling on surface fuel load estimation. Direct visual estimates often overestimated fuel loads where actual fuel loading was low and underestimated fuel loads where fuel loads were large. We found that double sampling with a classical regression estimation approach provided the most accurate and precise fuel load estimates, substantially improving the accuracy and precision achieved over standard photoload estimation when ń ≥ 20 and double sampling rate ≥20%. These results indicate that fine woody fuel loading estimation with the photoload technique can be improved by incorporating a double sampling approach.

Mixture of linear mixed models using multivariate t distribution

Linear mixed models are widely used when multiple correlated measurements are made on each unit of interest. In many applications, the units may form several distinct clusters, and such heterogeneity can be more appropriately modelled by a finite mixture linear mixed model. The classical estimation approach, in which both the random effects and the error parts are assumed to follow normal distribution, is sensitive to outliers, and failure to accommodate outliers may greatly jeopardize the model estimation and inference. We propose a new mixture linear mixed model using multivariate t distribution. For each mixture component, we assume the response and the random effects jointly follow a multivariate t distribution, to conveniently robustify the estimation procedure. An efficient expectation conditional maximization algorithm is developed for conducting maximum likelihood estimation. The degrees of freedom parameters of the t distributions are chosen data adaptively, for achieving flexible trade-off between estimation robustness and efficiency. Simulation studies and an application on analysing lung growth longitudinal data showcase the efficacy of the proposed approach.

Sequential Monte Carlo filter for state estimation of LiFePO4 batteries based on an online updated model

Battery state monitoring is one of the key techniques in battery management systems e.g. in electric vehicles. An accurate estimation can help to improve the system performance and to prolong the battery remaining useful life. Main challenges for the state estimation for LiFePO4 batteries are the flat characteristic of open-circuit-voltage over battery state of charge (SOC) and the existence of hysteresis phenomena. Classical estimation approaches like Kalman filtering show limitations to handle nonlinear and non-Gaussian error distribution problems. In addition, uncertainties in the battery model parameters must be taken into account to describe the battery degradation. In this paper, a novel model-based method combining a Sequential Monte Carlo filter with adaptive control to determine the cell SOC and its electric impedance is presented. The applicability of this dual estimator is verified using measurement data acquired from a commercial LiFePO4 cell. Due to a better handling of the hysteresis problem, results show the benefits of the proposed method against the estimation with an Extended Kalman filter.

Examining the Robustness of Competing Explanations of Slow Growth in African Countries

This research examines previous findings regarding the robustness of the African growth dummy by (i) expanding the list of variables to include those suggested by Easterly and Levine (1998) and Sachs and Warner (1997b) and (ii) with a modelling strategy that allows explicitly for model uncertainty. Using the Bayesian Averaging of Classical Estimates approach, this paper concludes that the African growth dummy does not appear to be robustly related to growth. This supports the interpretation that the significance of the African dummy may be attributable to omitted variable bias. The paper also contributes to the debate on growth strategies for Africa by assessing the robustness of divergent perspectives offered in the recent literature

Random effects in promotion time cure rate models

In this paper, a survival model with long-term survivors and random effects, based on the promotion time cure rate model formulation for models with a surviving fraction is investigated. We present Bayesian and classical estimation approaches. The Bayesian approach is implemented using a Markov chain Monte Carlo (MCMC) based on the Metropolis–Hastings algorithms. For the second one, we use restricted maximum likelihood (REML) estimators. A simulation study is performed to evaluate the accuracy of the applied techniques for the estimates and their standard deviations. An example on an oropharynx cancer study is used to illustrate the model and the estimation approaches considered in the study.

Why do some countries have a long‐term dependence on food aid?

PurposeThe purpose of this paper is to investigate the main determinants of dependency on food aid programmes. Food aid may take the form of a loan, a sale below market price or a current transfer. Food aid programmes across the globe are generally designed to provide short‐term assistance to countries. However, many countries have developed a dependence on food aid assistance.Design/methodology/approachThe paper estimates cross‐sectional regressions using a database containing 116 developing countries over the period 1970‐2003. Bayesian averaging of classical estimates is employed to identify the robust determinants of dependency.FindingsThe study finds that the top two determinants of food aid dependence (both cereals and non‐cereals) were food production and the frequency of droughts affecting the country. Food inflation, population density, crop yields, the amount of arable land per capita, the rule of law and the number of armed conflicts were also robustly related to aid dependence.Practical implicationsThese findings suggest that international donors should focus primarily on offsetting the substitution effect of aid on local production as well as implement systems to partially offset the negative effects of droughts, if they are to break the cycle of dependence on food aid.Originality/valuePrevious studies have investigated the effects of a large number of other variables. The findings from these papers were somewhat conflicting due to differences in model specification. To identify the variables that have a robust relationship with food aid dependence, whatever the model specification, the Bayesian averaging of classical estimates approach proposed by Sala‐i‐Martin et al. is employed.

Revisiting Aggregate U.S. Meat Demand with a Bayesian Averaging of Classical Estimates Approach: Do We Need a More General Theory?

AbstractAlthough meat demand is one of the most studied issues in agricultural economics, our understanding of this phenomenon has been hampered by valid concerns about model specification uncertainty. This article revisits the need for more general theories of aggregate U.S. meat demand. Using a Bayesian averaging of classical estimates approach, we draw comprehensive inferences over 1,048,576 demand systems. We find very little evidence supporting the need for more general theories that include demand determinants beyond prices and expenditures. We find strong evidence in support of symmetry and negativity, but strong evidence against homogeneity, which is consistent with other research.

Nonlinearities in cross-country growth regressions: A Bayesian Averaging of Thresholds (BAT) approach

We propose a Bayesian Averaging of Thresholds (BAT) approach for assessing the existence and quantifying the effect of threshold effects in cross-country growth regressions in the presence of model uncertainty. The BAT method extends the Bayesian Averaging of Classical Estimates (BACE) approach proposed by Sala-i-Martin et al. (2004) by allowing for uncertainty over nonlinear threshold effects. We apply our method to a set of determinants of long-term economic growth in a cross-section of 88 countries. Our results suggest that when model uncertainty is taken into account there is no evidence for robust threshold effects caused by the Initial Income, measured by GDP per capita in 1960, but that the Number of Years an Economy Has Been Open is an important source of nonlinear effects on growth.

AN ACCELERATED ALGORITHM FOR DENSITY ESTIMATION IN LARGE DATABASES USING GAUSSIAN MIXTURES

Today, with the advances of computer storage and technology, there are huge datasets available, offering an opportunity to extract valuable information. Probabilistic approaches are specially suited to learn from data by representing knowledge as density functions. In this paper, we choose Gaussian mixture models (GMMs) to represent densities, as they possess great flexibility to adequate to a wide class of problems. The classical estimation approach for GMMs corresponds to the iterative algorithm of expectation maximization (EM). This approach, however, does not scale properly to meet the high demanding processing requirements of large databases. In this paper we introduce an EM-based algorithm, that solves the scalability problem. Our approach is based on the concept of data condensation which, in addition to substantially diminishing the computational load, provides sound starting values that allow the algorithm to reach convergence faster. We also focus on the model selection problem. We test our algorithm using synthetic and real databases, and find several advantages, when compared to other standard existing procedures.

Mean-Squared Error Sampling and Reconstruction in the Presence of Noise

One of the main goals of sampling theory is to represent a continuous-time function by a discrete set of samples. Here, we treat the class of sampling problems in which the underlying function can be specified by a finite set of samples. Our problem is to reconstruct the signal from nonideal, noisy samples, which are modeled as the inner products of the signal with a set of sampling vectors, contaminated by noise. To mitigate the effect of the noise and the mismatch between the sampling and reconstruction vectors, the samples are linearly processed prior to reconstruction. Considering a statistical reconstruction framework, we characterize the strategies that are mean-squared error (MSE) admissible, meaning that they are not dominated in terms of MSE by any other linear reconstruction. We also present explicit designs of admissible reconstructions that dominate a given inadmissible method. Adapting several classical estimation approaches to our particular sampling problem, we suggest concrete admissible reconstruction methods and compare their performance. The results are then specialized to the case in which the samples are processed by a digital correction filter

Intelligence, Human Capital, and Economic Growth: A Bayesian Averaging of Classical Estimates (BACE) Approach

Human capital plays an important role in the theory of economic growth, but it has been difficult to measure this abstract concept. We survey the psychological literature on cross-cultural IQ tests and conclude that intelligence tests provide one useful measure of human capital. Using a new database of national average IQ, we show that in growth regressions that include only robust control variables, IQ is statistically significant in 99.8% of these 1330 regressions, easily passing a Bayesian model-averaging robustness test. A 1 point increase in a nation’s average IQ is associated with a persistent 0.11% annual increase in GDP per capita.

Intelligence, Human Capital, and Economic Growth: A Bayesian Averaging of Classical Estimates (BACE) Approach

Human capital plays an important role in the theory of economic growth, but it has been difficult to measure this abstract concept. We survey the psychological literature on cross-cultural IQ tests and conclude that modern intelligence tests provide one useful measure of human capital. Using a new database of national average IQ [Lynn and Vanhanen, 2002] along with a methodology derived from Sala-i-Martin, Doppelhofer, and Miller [AER, 2004], we show that in growth regressions that include only robust control variables, IQ is statistically significant at the 95% level in 99.8% of these 1330 regressions, and the coefficient is always positive. IQ readily passes a variety of robustness tests based upon Bayesian model averaging. A 1 point increase in a nation's average IQ is associated with a persistent 0.11% annual increase in GDP per capita.

Determinants of Long-Term Growth: A Bayesian Averaging of Classical Estimates (BACE) Approach

This paper examines the robustness of explanatory variables in cross-country economic growth regressions. It introduces and employs a novel approach, Bayesian Averaging of Classical Estimates (BACE), which constructs estimates by averaging OLS coefficients across models. The weights given to individual regressions have a Bayesian justification similar to the Schwarz model selection criterion. Of 67 explanatory variables we find 18 to be significantly and robustly partially correlated with long-term growth and another three variables to be marginally related. The strongest evidence is for the relative price of investment, primary school enrollment, and the initial level of real GDP per capita.

Spectral estimation based on the Wigner-Ville representation

Several classical techniques to estimate the component frequencies in a multicomponent signal are based on Fourier-like transformations. The frequency estimates obtained from their spectral peaks are affected by the window length and phase, thus presenting a large variance even in the absence of noise. Analytic signal based spectral estimators present no phase dependence for mono-component signals but, contrary to previous claims, they are not phase invariant for multicomponent signals, and perform worse than their real signal counterparts in high noise. The analysis of Wigner-Ville (WVD) spectra of continuous and discrete signals with time-limited windows demonstrates a better frequency concentration and less phase dependence than real or analytic signal Fourier spectra. The WVD presents accurate frequency estimates for multicomponent stationary signals, where crossterm interference is attenuated by smoothing the WVD in time (SWVD). It also has an excellent performance in the presence of noise, making it a good alternative to classical spectral estimation approaches. Furthermore, it is especially appropriate for the case of nonstationary multicomponent signals due to the good WVD temporal resolution, thus representing a superior spectral estimation technique suitable for the analysis of a variety of physical processes.

Model occurrence and model selection in panel data sets

In this study we focus attention on model selection in the presence of panel data. Our approach is eclectic in that it combines both classical and Bayesian techniques. It is also novel in that we address not only model selection, but also model occurrence, i.e., the process by which ‘nature’ chooses a statistical framework in which to generate the data of interest. For a given data subset, there exist competing models each of which have an ex ante positive probability of being the correct model, but for any one generated sample, ex post exactly one such model is the basis for the observed data set. Attention focuses on how the underlying model occurrence probabilities of the competing models depend on characteristics of the environments in which the data subsets are generated. Classical, Bayesian, and mixed estimation approaches are developed. Bayesian approaches are shown to be especially attractive whenever the models are nested.