Search

Multilevel models with nested errors are widely used in poverty estimation. An important application in this context is estimating the distribution of poverty as defined by the distribution of income within a set of domains that cover the population of interest. Since unit-level values of income are usually heteroskedastic, the standard homoskedasticity assumptions implicit in popular multilevel models may not be appropriate and can lead to bias, particularly when used to estimate domain-specific income distributions. This article addresses this problem when the income values in the population of interest can be characterized by a two-level mixed linear model with independent and identically distributed domain effects and with independent but not identically distributed individual effects. Estimation of poverty indicators that are functionals of domain-level income distributions is also addressed, and a nonparametric bootstrap procedure is used to estimate mean squared errors and confidence intervals. The proposed methodology is compared with the well-known World Bank poverty mapping methodology for this situation, using model-based simulation experiments as well as an empirical study based on Bangladesh poverty data. [ABSTRACT FROM AUTHOR]

Sampling is informative when probabilities of sample inclusion depend on unknown variables that are correlated with a response variable of interest. When sample inclusion probabilities are available, inverse probability weighting can be used to account for informative sampling in such a situation, although usually at the cost of less precise inference. This paper reviews two important research contributions by Chris Skinner that modify these weights to reduce their variability while at the same time retaining consistency of the weighted estimators. In some cases, however, sample inclusion probabilities are not known, and are estimated as propensity scores. This is often the situation in causal analysis, and double robust methods that protect against the resulting misspecification of the sampling process have been the focus of much recent research. In this paper we propose two model‐assisted modifications to the popular inverse propensity score weighted estimator of an average treatment effect, and then illustrate their use in a causal analysis of a rainfall enhancement experiment that was carried out in Oman between 2013 and 2018. [ABSTRACT FROM AUTHOR]

Summary: Although cloud seeding is a commonly used and plausible method for rainfall enhancement, its practical efficacy has not been established for seeding of convective clouds with hygroscopic materials. Other methods of rainfall enhancement are viewed as much less plausible. Thus, although increased electrical charge has been shown to enhance precipitation in cloud chamber experiments, exactly how ionisation of clouds can increase rainfall in the open atmosphere remains conjectural. A trial of the efficacy of ionisation for rainfall enhancement in the Hajar Mountains of Oman was carried out over 2013–2018. This paper provides some background to this non‐mainstream approach to increasing rainfall, showing how statistical modelling of rainfall data might be used to nudge rainfall enhancement via ionisation towards a more scientifically acceptable status. Analysis of the data collected in the trial shows that ionisation led to a statistically significant enhancement in positive rainfall in gauges located up to 70 km downwind of the ionisers. A headline analysis specified prior to commencement of the trial resulted in an estimate of 16.23% enhancement relative to rainfall that would have fallen without any ionisation, while a more sophisticated after the event analysis increased this estimate to 17.64%. [ABSTRACT FROM AUTHOR]

Summary: Applications that use linked data are now part of mainstream social science research, though they generally do not take linkage error into consideration. Solutions that correct for the bias caused by these errors have been proposed but are not yet embedded in the various analysis procedures in common use. Secondary analyses based on linked data can therefore be potentially misleading. We review some recent approaches to non‐deterministic data linkage together with a framework for secondary analysis of the linked data which makes use of paradata produced by the linkage process to correct this bias. We also describe a new method for secondary analysis of linked data that builds on this framework and show how it can be used for estimation of a set of domain means based on linked data. We then illustrate this approach via an empirical study based on record linkage of agricultural producers in four states of Brazil aimed at producing estimates of agricultural output by industry. Our study considers register‐to‐register linkage as well as sample‐to‐register linkage, and we show results for the traditional Fellegi–Sunter approach to record linkage as well as for a newer linkage procedure based on the use of classification trees and bagging. [ABSTRACT FROM AUTHOR]

Model-based and model-assisted methods of survey estimation aim to improve the precision of estimators of the population total or mean relative to methods based on the nonparametric Horvitz-Thompson estimator. These methods often use a linear regression model defined in terms of auxiliary variables whose values are assumed known for all population units. Information on networks represents another form of auxiliary information that might increase the precision of these estimators, particularly if it is reasonable to assume that networked population units have similar values of the survey variable. Linear models that use networks as a source of auxiliary information include autocorrelation, disturbance, and contextual models. In this article we focus on social networks, and investigate how much of the population structure of the network needs to be known for estimation methods based on these models to be useful. In particular, we use simulation to compare the performance of the best linear unbiased predictor under a model that ignores the network with model-based estimators that incorporate network information. Our results show that incorporating network information via a contextual model seems to be the most appropriate approach. We also show that one does not need to know the full population network, but that knowledge of the partial network linking the sampled population units to the non-sampled population units is necessary. Finally, we also provide an estimator for the mean-squared error to make an informed decision about using the contextual information, as well as the results showing that this adaptive strategy leads to higher precision. [ABSTRACT FROM AUTHOR]

Oman is one of most water-stressed countries in the world, faced with declining reserves of fresh groundwater and the prospect of decreased natural rainfall due to climate change. A technology that can increase natural rainfall and mitigate the current and future imbalance between Oman's demand and availability of water resources is of major interest. A rainfall enhancement trial using a ground-based ionisation technology was conducted in the Hajar Mountains in western Oman over a period of 170 days (May to October) in 2013. Data for the trial were obtained from 120 rain gauges and two automatic weather stations installed for the trial, as well as data from 11 weather stations operated by the Oman Directorate General of Meteorology and Air Navigation and the daily upper air sounding at Muscat International Airport. The trial employed a crossover design with the two ion emitters operated in a pre-determined randomised alternating schedule. Statistical analysis of the trial data was carried out using spatio-temporal models that used meteorological and spatio-temporal covariates to capture natural rainfall variation, enabling prediction of the level of rainfall that would have occurred if the ionisation emitters had not been operated. This methodology used dynamically defined target and control areas corresponding to two 30-km wide 'corridors' placed symmetrically about each emitter site and oriented downwind along the axis defined by the steering wind direction. Overall, a positive and significant rainfall enhancement effect attributable to the operation of the emitters was observed. The trial was continued in 2014, expanding the trial area to include four ion emitters and 149 rain gauges. Results from this year are also discussed. [ABSTRACT FROM AUTHOR]

Survey data often contain measurements for variables that are semicontinuous in nature, i.e. they either take a single fixed value (we assume this is zero) or they have a continuous, often skewed, distribution on the positive real line. Standard methods for small area estimation (SAE) based on the use of linear mixed models can be inefficient for such variables. We discuss SAE techniques for semicontinuous variables under a two part random effects model that allows for the presence of excess zeros as well as the skewed nature of the nonzero values of the response variable. In particular, we first model the excess zeros via a generalized linear mixed model fitted to the probability of a nonzero, i.e. strictly positive, value being observed, and then model the response, given that it is strictly positive, using a linear mixed model fitted on the logarithmic scale. Empirical results suggest that the proposed method leads to efficient small area estimates for semicontinuous data of this type. We also propose a parametric bootstrap method to estimate the MSE of the proposed small area estimator. These bootstrap estimates of the MSE are compared to the true MSE in a simulation study. [ABSTRACT FROM AUTHOR]

The author comments on the articles by J. Burger et al, S. Gerritse et al and T. Tuto and L. Di Consiglio which tackle measurement-error issues that are important emerging features of official statistics methodology.Topics discusses include their common focus on problems that arise when situations corresponding to nonstandard measurement error scenarios come up in official statistics and the need to apply model-based inference paradigm for official statistics with a strong dose of common sense.

A new semiparametric approach to model-based small area prediction for counts is proposed and used for estimating the average number of visits to physicians for Health Districts in Central Italy. The proposed small area predictor can be viewed as an outlier robust alternative to the more commonly used empirical plug-in predictor that is based on a Poisson generalized linear mixed model with Gaussian random effects. Results from the real data application and from a simulation experiment confirm that the proposed small area predictor has good robustness properties and in some cases can be more efficient than alternative small area approaches. [ABSTRACT FROM PUBLISHER]

Small area estimates based on the widely used area-level model proposed in Fay and Herriot (1979) assume that the area-level direct estimates are spatially uncorrelated. In many cases, however, this is not the case. Extensions of the Fay-Herriot model to allow for spatial correlation have been proposed, but all assume spatial stationarity; i.e., the parameters of the associated regression model for the small area characteristic of interest do not vary spatially. Instead, spatial effects are introduced by imposing a spatial correlation structure on the regression errors. In this paper, we propose an extension to the Fay-Herriot model that accounts for the presence of spatial nonstationarity, i.e., where the parameters of this regression model vary spatially. We refer to the predictor based on this extended model as the nonstationary empirical best linear unbiased predictor (NSEBLUP). We also develop two different estimators for the mean squared error of the NSEBLUP. The first estimator uses approximations similar to those in Opsomer, Claeskens, Ranalli, Kauermann, and Breidt (2008). The second estimator is based on the parametric bootstrapping approach of Gonzalez-Manteiga, Lombardia, Molina, Morales, and Santamaria (2008) and Molina, Salvati, and Pratesi (2009). Results from model-based and design-based simulation studies using spatially nonstationary data indicate that the NSEBLUP compares favorably with alternative area-level predictors that ignore this spatial nonstationarity. In addition, both proposed methods for estimating its mean squared error seem adequate. [ABSTRACT FROM AUTHOR]

We introduce a semi-parametric approach to ecological regression for disease mapping, based on modelling the regression M-quantiles of a negative binomial variable. The proposed method is robust to outliers in the model covariates, including those due to measurement error, and can account for both spatial heterogeneity and spatial clustering. A simulation experiment based on the well-known Scottish lip cancer data set is used to compare the M-quantile modelling approach with a disease mapping approach based on a random effects model. This suggests that the M-quantile approach leads to predicted relative risks with smaller root mean square error. The paper concludes with an illustrative application of the M-quantile approach, mapping low birth weight incidence data for English Local Authority Districts for the years 2005-2010. Copyright © 2014 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Probabilistic matching of records is widely used to create linked data sets for use in health science, epidemiological, economic, demographic and sociological research. Clearly, this type of matching can lead to linkage errors, which in turn can lead to bias and increased variability when standard statistical estimation techniques are used with the linked data. In this paper we develop unbiased regression parameter estimates to be used when fitting a linear model with nested errors to probabilistically linked data. Since estimation of variance components is typically an important objective when fitting such a model, we also develop appropriate modifications to standard methods of variance components estimation in order to account for linkage error. In particular, we focus on three widely used methods of variance components estimation: analysis of variance, maximum likelihood and restricted maximum likelihood. Simulation results show that our estimators perform reasonably well when compared to standard estimation methods that ignore linkage errors. [ABSTRACT FROM AUTHOR]

Binary data are often of interest in business surveys, particularly when the aim is to characterize grouping in the businesses making up the survey population. When small area estimates are required for such binary data, use of standard estimation methods based on linear mixed models (LMMs) becomes problematic. We explore two model-based techniques of small area estimation for small area proportions, the empirical best predictor (EBP) under a generalized linear mixed model and the model-based direct estimator (MBDE) under a population-level LMM. Our empirical results show that both the MBDE and the EBP perform well. The EBP is a computationally intensive method, whereas the MBDE is easy to implement. In case of model misspecification, the MBDE also appears to be more robust. The mean-squared error (MSE) estimation of MBDE is simple and straightforward, which is in contrast to the complicated MSE estimation for the EBP. [ABSTRACT FROM AUTHOR]

An additional field trial of an ionisation technology, called Atlant, was conducted in late 2010. Previous analysis of the data collected in field trials of the technology conducted in 2008 and 2009 used spatio-temporal statistical models to account for the impact of meteorological and topographic conditions not controllable by the randomised experimental design. In addition, a novel application of a random effect block bootstrap was developed for inference. These techniques are applied to the analysis of the 2010 Atlant field trial. In response to peer-review of previous trials, a new modelling approach is developed for this 2010 analysis, that uses dynamically defined upwind control areas to generate values of an instrumental variable that integrates the effects of meteorology and topography induced variation in rainfall. This allows a much simpler model specification and a clearer delineation between naturally occurring rainfall and any additional rainfall attributable to the operation of Atlant. Results using both the statistical methodology of previous trials and also by fitting this so-called "instrumental" model are consistent with those obtained in previous analyses, which had suggested a positive increases in rainfall of around nine percent relative to the predicted rainfall that would have occurred in the absence of Atlant operation. [ABSTRACT FROM AUTHOR]

Much of the small-area estimation literature focuses on population totals and means. However, users of survey data are often interested in the finite-population distribution of a survey variable and in the measures (e.g. medians, quartiles, percentiles) that characterize the shape of this distribution at the small-area level. In this paper we propose a model-based direct estimator (MBDE, Chandra and Chambers) of the small-area distribution function. The MBDE is defined as a weighted sum of sample data from the area of interest, with weights derived from the calibrated spline-based estimate of the finite-population distribution function introduced by Harms and Duchesne, under an appropriately specified regression model with random area effects. We also discuss the mean squared error estimation of the MBDE. Monte Carlo simulations based on both simulated and real data sets show that the proposed MBDE and its associated mean squared error estimator perform well when compared with alternative estimators of the area-specific finite-population distribution function. [ABSTRACT FROM AUTHOR]

Weather modification trials tend to rely on randomized experimental designs. Unfortunately, these designs have so far not demonstrated sufficient power to detect a small weather modification signal againstthe large level of background variation in rainfall. Further, randomized experimental designs are generally not possible when dealing with ground-based sources of weather modification such as industrial pollution. Statistical modeling of rainfall gauge measurements that attempt to control for meteorological and orographic variation in rainfall measurements seem better suited in this regard. Evaluation would be relatively simple if we could separate the sources of variation into changes in meteorological conditions in time and fixed effects due to the location of rainfall gauges. Unfortunately, a large part of the natural variation in rainfall measurements is caused by a mix of spatial and temporal influences. Meteorological conditions are not spatially homogeneous and orographic effects can depend on prevailing conditions. Importantly, exposure of rainfall gauges to an effect is generally dependent on meteorological conditions, primarily wind direction and speed. A ground-based rainfall enhancement trial was conducted using a randomized crossover design in South Australia in 2009. The analysis presented in this paper explores the limitations imposed by ignoring spatiotemporal variation in the rainfall data collected in this trial and takes advantage of modern statistical methods to construct an appropriately specified model for these data. The level at which analysis is performed is addressed, particularly whether it is appropriate in this situation to use gauge-level data, as opposed to aggregated data such as average daily rainfall, in statistical inference. Our analysis, which accounts for the spatial and temporal correlation structure of rainfall data, suggests that there is a substantial increase in the likelihood of detecting a modification signal when the analysis is carried out at the gauge level. [ABSTRACT FROM AUTHOR]

Small-area estimation techniques have typically relied on plug-in estimation based on models containing random area effects. More recently, regression M-quantiles have been suggested for this purpose, thus avoiding conventional Gaussian assumptions, as well as problems associated with the specification of random effects. However, the plug-in M-quantile estimator for the small-area mean can be shown to be the expected value of this mean with respect to a generally biased estimator of the small-area cumulative distribution function of the characteristic of interest. To correct this problem, we propose a general framework for robust small-area estimation, based on representing a small-area estimator as a functional of a predictor of this small-area cumulative distribution function. Key advantages of this framework are that it naturally leads to integrated estimation of small-area means and quantiles and is not restricted to M-quantile models. We also discuss mean squared error estimation for the resulting estimators, and demonstrate the advantages of our approach through model-based and design-based simulations, with the latter using economic data collected in an Australian farm survey. [ABSTRACT FROM AUTHOR]

Non-stationary spatial variation makes it difficult to establish real-time areas of control and effect in weather modification. Non-stationary temporal variation makes the comparison of long-term averages from limited climatic records open to question. Here we describe a statistical methodology which addresses both problems explicitly, in a trial of a ground-based ionization technology known as Atlant, and which could be applied to other weather modification technologies more generally. The approach adopted here is based on a statistical model for daily rainfall that achieves a high level of real-time control by the inclusion of both spatial and temporal components. In particular, it makes use of daily gauge level rainfall data, orographic and daily meteorological covariates, as well as dynamically defined downwind areas, to model the impact of Atlant operation on rainfall. Subject to the caveat that the trial was not randomized in any way, this type of dynamic control demonstrates a clear rainfall enhancement effect at both a simple observational level and when a spatial random effects model is used to control for covariates. Rainfall downwind of the Atlant test site was 15% higher than rainfall in the control (crosswind or upwind) areas. Based on these results, randomized trials with multiple sites are currently being conducted in the same area. [ABSTRACT FROM AUTHOR]

Count response data often exhibit departures from the assumptions of standard Poisson generalized linear models. In particular, cluster level correlation of the data and truncation at zero are two common characteristics of such data. This paper describes a random components truncated Poisson model that can be applied to clustered and zero-truncated count data. Residual maximum likelihood method estimators for the parameters of this model are developed and their use is illustrated using a dataset of non-zero counts of sheets with edge-strain defects in iron sheets produced by the Mobarekeh Steel Complex, Iran. The paper also reports on a small-scale simulation study that supports the estimation procedure. [ABSTRACT FROM AUTHOR]

Discusses the paper presented by Jon Wakefield to the "Journal of the Royal Statistical Society," regarding the careful modelling of ecological data. Argument for the appropriateness of the binomial likelihood, in many settings; Types of contextual effects allowed by the model presented by Wakefield; Discussion on the connection with ecological regression carried out in spatial epidemiology outlined in a section of his paper.