Misspecification Of Distribution Research Articles

Model-implied simulation-based power estimation for correctly specified and distributionally misspecified models: Applications to nonlinear and linear structural equation models.

Closed-form (asymptotic) analytical power estimation is only available for limited classes of models, requiring correct model specification for most applications. Simulation-based power estimation can be applied in almost all scenarios where data following the model can be estimated. However, a general framework for calculating the required sample sizes for given power rates is still lacking. We propose a new model-implied simulation-based power estimation (MSPE) method for the z-test that makes use of the asymptotic normality property of estimates of a wide class of estimators, the M-estimators, and give theoretical justification for the approach. M-estimators include maximum-likelihood, least squares estimates and limited information estimators, but also estimators used for misspecified models, hence, the new simulation-based power modeling method is widely applicable. The MSPE employs a parametric model to describe the relationship between power and sample size, which can then be used to determine the required sample size for a specified power rate. We highlight its performance in linear and nonlinear structural equation models (SEM) for correctly specified models and models under distributional misspecification. Simulation results suggest that the new power modeling method is unbiased and shows good performance with regard to root mean squared error and type I error rates for the predicted required sample sizes and predicted power rates, outperforming alternative approaches, such as the naïve approach of selecting a discrete selection of sample sizes with linear interpolation of power or simple logistic regression approaches. The MSPE appears to be a valuable tool to estimate power for models without an (asymptotic) analytical power estimation.

Bootstrap-based statistical inference for linear mixed effects under misspecifications

Linear mixed effects are considered excellent predictors of cluster-level parameters in various domains. However, previous research has demonstrated that their performance is affected by departures from model assumptions. Given the common occurrence of these departures in empirical studies, there is a need for inferential methods that are robust to misspecifications while remaining accessible and appealing to practitioners. Statistical tools have been developed for cluster-wise and simultaneous inference for mixed effects under distributional misspecifications, employing a user-friendly semiparametric random effect bootstrap. The merits and limitations of this approach are discussed in the general context of model misspecification. Theoretical analysis demonstrates the asymptotic consistency of the methods under general regularity conditions. Simulations show that the proposed intervals are robust to departures from modelling assumptions, including asymmetry and long tails in the distributions of errors and random effects, outperforming competitors in terms of empirical coverage probability. Finally, the methodology is applied to construct confidence intervals for household income across counties in the Spanish region of Galicia.

Identification and estimation of panel semiparametric conditional heteroskedastic frontiers with dynamic inefficiency

We study a semiparametric panel stochastic frontier model with nonseparable unobserved heterogeneity, which allows for time-varying conditional heteroskedastic productivity components. It does not require distributional assumptions on random noise except conditional symmetry. We utilize conditional characteristic functions from Kotlarski’s Lemma to derive new moment conditions that yield the identification of the heteroskedastic variance parameters of inefficiency and random noise. Identification only requires a panel with three periods for serially correlated inefficiency. A nonparametric estimation procedure is also developed for the conditional variance of inefficiency, and its convergence rate is established. Monte Carlo simulation shows that the estimator is robust to misspecification of inefficiency distributions.

A New Paradigm for High-dimensional Data: Distance-Based Semiparametric Feature Aggregation Framework via Between-Subject Attributes.

This article proposes a distance-based framework incentivized by the paradigm shift towards feature aggregation for high-dimensional data, which does not rely on the sparse-feature assumption or the permutation-based inference. Focusing on distance-based outcomes that preserve information without truncating any features, a class of semiparametric regression has been developed, which encapsulates multiple sources of high-dimensional variables using pairwise outcomes of between-subject attributes. Further, we propose a strategy to address the interlocking correlations among pairs via the U-statistics-based estimating equations (UGEE), which correspond to their unique efficient influence function (EIF). Hence, the resulting semiparametric estimators are robust to distributional misspecification while enjoying root-n consistency and asymptotic optimality to facilitate inference. In essence, the proposed approach not only circumvents information loss due to feature selection but also improves the model's interpretability and computational feasibility. Simulation studies and applications to the human microbiome and wearables data are provided, where the feature dimensions are tens of thousands.

A self-censoring model for multivariate nonignorable nonmonotone missing data.

We introduce an itemwise modeling approach called "self-censoring" for multivariate nonignorable nonmonotone missing data, where the missingness process of each outcome can be affected by its own value and associated with missingness indicators of other outcomes, while conditionally independent of the other outcomes. The self-censoring model complements previous graphical approaches for the analysis of multivariate nonignorable missing data. It is identified under a completeness condition stating that any variability in one outcome can be captured by variability in the other outcomes among complete cases. For estimation, we propose a suite of semiparametric estimators including doubly robust estimators that deliver valid inferences under partial misspecification of the full-data distribution. We also provide a novel and flexible global sensitivity analysis procedure anchored at the self-censoring. We evaluate the performance of the proposed methods with simulations and apply them to analyze a study about the effect of highly active antiretroviral therapy on preterm delivery of HIV-positivemothers.

The Role of Ambiguity Aversion in Contract-farming Supply Chains: A Distributionally Robust Approach

Contract farming is an important tool to promote the agricultural development in the developing countries. In this paper, we study three types of contracts commonly used in a contract-farming agricultural supply chain with a farmer and an agribusiness firm, including fixed procurement price contract, market spot price contract and protection procurement price contract. To accurately capture the correlation between uncertainties and avoid the misspecification of true distribution, a moment-based distributionally robust optimization framework is introduced to analyze the three contracts. We then derive equilibrium solutions of both the farmer and the agribusiness firm under each price contract. Leveraging these results, we can prove that the protection procurement price contract is equivalent to either the fixed procurement price contract or the market spot price contract, which depends on under which contract the farmer can obtain a larger marginal revenue. We further investigate the intrinsic mechanism of the protection procurement price contract, including its impact on the performance of both contract parties and the entire supply chain. Finally, we provide numerical comparisons of our ambiguity-averse solution with the ambiguity-neutral one and also discuss the impact of demand and yield unertainties on the equilibrium profits of both contract parties.

Robust quantile regression analysis for probabilistic modelling of S-N curves

The scatter in fatigue data is commonly characterized by probability distributions for constructing the probabilistic S-N curves. However, there is notable estimation bias under distribution misspecification. In this paper, we proposed a quantile regression framework for modeling S-N curves. The quantile regression model can be built directly on the experimental data without any distribution assumption. Extensive simulations and two experimental datasets are used to illustrate the usefulness of the proposed model. The results demonstrate that the quantile regression model is exempt from the problem of incorrectly specifying the potential fatigue life distribution and is robust to the non-constant scale problem.

A 2-stage elastic net algorithm for estimation of sparse networks with heavy-tailed data

We propose a new 2-stage procedure that relies on the elastic net penalty to estimate a network based on partial correlations when data are heavy-tailed. The new estimator allows us to consider the LASSO penalty as a special case. Extensive simulation analysis shows that the 2-stage estimator performs best for heavy-tailed data and it is also robust to distribution misspecification, both in terms of identification of the sparsity patterns and numerical accuracy. Empirical results on real-world data focus on the estimation of the European banking network during the Covid-19 pandemic. We show that the new estimator can provide interesting insights both for the development of network indicators, such as network strength, to identify crisis periods and for the detection of banking network properties, such as centrality and level of interconnectedness, that might play a relevant role in setting up adequate risk management and mitigation tools.

Prediction of random effects with misspecified AFT random effect models

The accelerated failure time (AFT) model with random effects has been widely used for clustered or correlated time-to-event data as an alternative to frailty model which is the Cox’s proportional hazards model with random effects. The AFT random effect model usually assumes a normal distribution for random effect distribution. It is well known that the estimated regression parameters in the AFT model are robust against various violations of the assumed model. However, the impact of prediction (or estimation) of random effect, when the assumed normal random effect is misspecified, has been relatively less studied. In this paper, we investigate the impact of misspecification of normal random effect distribution on the prediction of random effect under the AFT random effect model. Here, the random effect is estimated using the hierarchical likelihood (h-likelihood) which is useful for the inference of random effects. The proposed method is demonstrated using simulation studies and a real data set.

Distributionally robust portfolio optimization with second-order stochastic dominance based on wasserstein metric

In portfolio optimization, we may be dealing with misspecification of a known distribution, that stock returns follow it. The unknown true distribution is considered in terms of a Wasserstein-neighborhood of P to examine the tractable formulations of the portfolio selection problem. This study considers a distributionally robust portfolio optimization problem with an ambiguous stochastic dominance constraint by assuming the unknown distribution of asset returns. The objective is to maximize the worst-case expected return and subject to an ambiguous second-order stochastic dominance constraint. The expected return robustly stochastically dominates the benchmark in the second order over all possible distributions within an ambiguity set. It is also shown that the Wasserstein-moment ambiguity set-based distributionally robust portfolio optimization can be reduced to a semidefinite program and second-order conic programming. We use a cutting plane to solve our second-order stochastic dominance constraint portfolio optimization problem with ambiguity sets based on the Wasserstein metric. Then we decompose this class of distributionally robust portfolio optimization into semi-infinite programming and apply the cutting surface method to solve it. The captured optimization programs are applied to real-life data for more efficient comparison. The problems are examined in depth using the optimal solutions of the optimization programs based on the different setups.

Beta distribution misspecification tests with application to Covid-19 mortality rates in the United States.

The beta distribution is routinely used to model variables that assume values in the standard unit interval, (0, 1). Several alternative laws have, nonetheless, been proposed in the literature, such as the Kumaraswamy and simplex distributions. A natural and empirically motivated question is: does the beta law provide an adequate representation for a given dataset? We test the null hypothesis that the beta model is correctly specified against the alternative hypothesis that it does not provide an adequate data fit. Our tests are based on the information matrix equality, which only holds when the model is correctly specified. They are thus sensitive to model misspecification. Simulation evidence shows that the tests perform well, especially when coupled with bootstrap resampling. We model state and county Covid-19 mortality rates in the United States. The misspecification tests indicate that the beta law successfully represents Covid-19 death rates when they are computed using either data from prior to the start of the vaccination campaign or data collected when such a campaign was under way. In the latter case, the beta law is only accepted when the negative impact of vaccination reach on death rates is moderate. The beta model is rejected under data heterogeneity, i.e., when mortality rates are computed using information gathered during both time periods.

Reliability evaluation of repairable systems considering component heterogeneity using frailty model

The failure processes of heterogeneous repairable systems with minimal repair assumption can be modelled by nonhomogeneous Poisson processes. One approach to describe an unobserved heterogeneity between systems is to multiply the intensity function by a positive random variable (frailty term) with a gamma distribution. This approach assumes that the relative frailty distribution among survivors is independent of age. Where systems are being continuously repaired and modified, the frailty distribution may be dependent on the system’s age. This paper investigates the application of the inverse Gaussian (IG) frailty model for modelling the failure processes of heterogeneous repairable systems. The IG frailty model, which combines the power law model and inverse Gaussian distribution, assumes that the relative frailty distribution among survivors becomes increasingly homogeneous over time. We develop the maximum likelihood for the IG frailty model, a method for event prediction, and investigate the effect of accuracy of the IG estimator and mis-specification of the frailty distribution through a simulation study. The mean estimates of the scale and shape parameters of the intensity function are examined for bias and efficiency loss. We find that the developed estimator is robust to changes in the input parameters for a relatively large sample sizes. We investigate the robustness of selecting an IG compared with a gamma frailty model. The developed IG model is applied to real data for illustration showing an improvement on existing models.

Discrete mixtures of normals pseudo maximum likelihood estimators of structural vector autoregressions

Likelihood inference in structural vector autoregressions with independent non-Gaussian shocks leads to parametric identification and efficient estimation at the risk of inconsistencies under distributional misspecification. We prove that autoregressive coefficients and (scaled) impact multipliers remain consistent, but the drifts and shocks’ standard deviations are generally inconsistent. Nevertheless, we show consistency when the non-Gaussian log-likelihood uses a discrete scale mixture of normals in the symmetric case, or an unrestricted finite mixture more generally, and compare the efficiency of these estimators to other consistent two-step proposals, including our own. Finally, our empirical application looks at dynamic linkages between three popular volatility indices.

Zero and One Inflated Item Response Theory Models for Bounded Continuous Data

Bounded continuous data are encountered in many applications of item response theory, including the measurement of mood, personality, and response times and in the analyses of summed item scores. Although different item response theory models exist to analyze such bounded continuous data, most models assume the data to be in an open interval and cannot accommodate data in a closed interval. As a result, ad hoc transformations are needed to prevent scores on the bounds of the observed variables. To motivate the present study, we demonstrate in real and simulated data that this practice of fitting open interval models to closed interval data can majorly affect parameter estimates even in cases with only 5% of the responses on one of the bounds of the observed variables. To address this problem, we propose a zero and one inflated item response theory modeling framework for bounded continuous responses in the closed interval. We illustrate how four existing models for bounded responses from the literature can be accommodated in the framework. The resulting zero and one inflated item response theory models are studied in a simulation study and a real data application to investigate parameter recovery, model fit, and the consequences of fitting the incorrect distribution to the data. We find that neglecting the bounded nature of the data biases parameters and that misspecification of the exact distribution may affect the results depending on the data generating model.

Estimation of vaccine efficacy for variants that emerge afterthe placebo group is vaccinated.

SARS‐CoV‐2 continues to evolve and the vaccine efficacy against variants is challenging to estimate. It is now common in phase III vaccine trials to provide vaccine to those randomized to placebo once efficacy has been demonstrated, precluding a direct assessment of placebo controlled vaccine efficacy after placebo vaccination. In this work, we extend methods developed for estimating vaccine efficacy post placebo vaccination to allow variant specific time varying vaccine efficacy, where time is measured since vaccination. The key idea is to infer counterfactual strain specific placebo case counts by using surveillance data that provide the proportions of the different strains. This blending of clinical trial and observational data allows estimation of strain‐specific time varying vaccine efficacy, or sieve effects, including for strains that emerge after placebo vaccination. The key requirements are that the surveillance strain distribution accurately reflects the strain distribution for a placebo group throughout follow‐up after placebo group vaccination, and that at least one strain is present before and after placebo vaccination. For illustration, we develop a Poisson approach for an idealized design under a rare disease assumption and then use a proportional hazards model to address staggered entry, staggered crossover, and smoothly varying strain specific vaccine efficacy. We evaluate these methods by theoretical work and simulations, and demonstrate that useful estimation of the efficacy profile is possible for strains that emerge after vaccination of the placebo group. An important principle is to incorporate sensitivity analyses to guard against misspecification of the strain distribution.

Goodness of Fit Test Based on BLUS Residuals for Error Distribution of Regression Model

The error distribution testing plays an important role in linear regression as distribution misspecification seriously affects the validity and efficiency of regression analysis. The least squares (OLS) residuals are often used to construct test statistics; in order to overcome the non-independent and identically residuals, the best linear unbiased scale (BLUS) residuals are applied in this paper, which, unlike OLS residuals, the residuals vector is identically and independently distributed. Based on the BLUS residuals, a new test statistic is constructed by using the sample random distance between sample quantile and quasi sample quantile derived from the null distribution, and the goodness-of-fit test of error distribution in the linear model is studied. The powers of the new tests under certain alternatives are examined. They are more powerful tests for the hypotheses concerned.

Multivariate generalized linear mixed models for continuous bounded outcomes: Analyzing the body fat percentage data.

We propose a multivariate regression model to handle multiple continuous bounded outcomes. We adopted the maximum likelihood approach for parameter estimation and inference. The model is specified by the product of univariate probability distributions and the correlation between the response variables is obtained through the correlation matrix of the random intercepts. For modeling continuous bounded variables on the interval we considered the beta and unit gamma distributions. The main advantage of the proposed model is that we can easily combine different marginal distributions for the response variable vector. The computational implementation is performed using Template Model Builder, which combines the Laplace approximation with automatic differentiation. Therefore, the proposed approach allows us to estimate the model parameters quickly and efficiently. We conducted a simulation study to evaluate the computational implementation and the properties of the maximum likelihood estimators under different scenarios. Moreover, we investigate the impact of distribution misspecification in the proposed model. Our model was motivated by a data set with multiple continuous bounded outcomes, which refer to the body fat percentage measured at five regions of the body. Simulation studies and data analysis showed that the proposed model provides a general and rich framework to deal with multiple continuous bounded outcomes.

Posterior Average Effects

Economists are often interested in estimating averages with respect to distributions of unobservables, such as moments of individual fixed-effects, or average partial effects in discrete choice models. For such quantities, we propose and study posterior average effects (PAE), where the average is computed conditional on the sample, in the spirit of empirical Bayes and shrinkage methods. While the usefulness of shrinkage for prediction is well-understood, a justification of posterior conditioning to estimate population averages is currently lacking. We show that PAE have minimum worst-case specification error under various forms of misspecification of the parametric distribution of unobservables. In addition, we introduce a measure of informativeness of the posterior conditioning, which quantifies the worst-case specification error of PAE relative to parametric model-based estimators. As illustrations, we report PAE estimates of distributions of neighborhood effects in the U.S., and of permanent and transitory components in a model of income dynamics.

Testing Error Distribution by Kernelized Stein Discrepancy in Multivariate Time Series Models

Knowing the error distribution is important in many multivariate time series applications. To alleviate the risk of error distribution mis-specification, testing methodologies are needed to detect whether the chosen error distribution is correct. However, the majority of existing tests only deal with the multivariate normal distribution for some special multivariate time series models, and thus cannot be used for testing the often observed heavy-tailed and skewed error distributions in applications. In this article, we construct a new consistent test for general multivariate time series models, based on the kernelized Stein discrepancy. To account for the estimation uncertainty and unobserved initial values, a bootstrap method is provided to calculate the critical values. Our new test is easy-to-implement for a large scope of multivariate error distributions, and its importance is illustrated by simulated and real data. As an extension, we also show how to test for the error distribution in copula time series models.

Relaxing conditional independence in an endogenous binary response model

For binary response models, the literature primarily addresses endogeneity by a control function approach assuming conditional independence (CF-CI). However, as the literature also notes, CF-CI implies conditions like homoskedasticity (of the latent error with respect to the instruments) that fail in many empirical settings. I propose an alternative approach that allows for heteroskedasticity, achieving identification with a conditional mean restriction. These identification results apply to a latent Gaussian error term with flexibly parametrized heteroskedasticity. I propose a two step conditional maximum likelihood estimator and derive its asymptotic distribution. In simulations, the new estimator outperforms others when CF-CI fails and is fairly robust to distributional misspecification.