Restricted Likelihood Ratio Test Research Articles

Exact testing for heteroscedasticity in a two‐way layout in variety frost trials when incorporating a covariate

SummaryTwo‐way layouts are common in grain industry research where it is often the case that there are one or more covariates. It is widely recognised that when estimating fixed effect parameters, one should also examine for possible extra error variance structure. An exact test for heteroscedasticity, when there is a covariate, is illustrated for a data set from frost trials in Western Australia. While the general algebra for the test is known, albeit in past literature, there are computational aspects of implementing the test for the two way when there are covariates. In this scenario the test is shown to have greater power than the industry standard, and because of its exact size, is preferable to use of the restricted maximum likelihood ratio test (REMLRT) based on the approximate asymptotic distribution in this instance. Formulation of the exact test considered here involves creation of appropriate contrasts in the experimental design. This is illustrated using specific choices of observations corresponding to an index set in the linear model for the two‐way layout. Also an algorithm supplied complements the test. Comparisons of size and power then ensue. The test has natural extensions when there are unbalanced data, and more than one covariate may be present. Results can be extended to Balanced Incomplete Block Designs.

On asymptotic distributions of several test statistics for familial relatedness in linear mixed models.

In this study, the asymptotic distributions of the likelihood ratio test (LRT), the restricted likelihood ratio test (RLRT), the F and the sequence kernel association test (SKAT) statistics for testing an additive effect of the expected familial relatedness (FR) in a linear mixed model are examined based on an eigenvalue approach. First, the covariance structure for modeling the FR effect in a LMM is presented. Then, the multiplicity of eigenvalues for the log-likelihood and restricted log-likelihood is established under a replicate family setting and extended to a more general replicate family setting (GRFS) as well. After that, the asymptotic null distributions of LRT, RLRT, F and SKAT statistics under GRFS are derived. The asymptotic null distribution of SKAT for testing genetic rare variants is also constructed. In addition, a simple formula for sample size calculation is provided based on the restricted maximum likelihood estimate of the effect size for the expected FR. Finally, a power comparison of these test statistics on hypothesis test of the expected FR effect is made via simulation. The four test statistics are also applied to a data set from the UK Biobank.

How standard are standardized drought indices? Uncertainty components for the SPI & SPEI case

Standardized drought indices such as the Standardized Precipitation Index (SPI) or the Standardized Precipitation and Evapotranspiration Index (SPEI) are frequently used around the world to assess drought severity across a continent or a larger region covering different meteorological regimes. But how standard are these standardized indices? In this paper we quantify the uncertainty of SPI and SPEI based on an Austrian dataset to shed light on what are the main sources of uncertainty in the study area. Five factors that either defy the control of the analyst (record length, observation period), or need to be subjectively decided during the steps of the calculation (choice of the distribution, parameter estimation method, and GOF-test of the fitted distribution) are considered. We use the root mean squared error ( E R M S ) for estimating the typical error for different calculation algorithm of SPI and SPEI. The total and relative uncertainty components for each factor are analysed by a linear mixed model (LMM) and significance of each model parameter are tested by the Akaike information criterion (AIC) and the restricted likelihood ratio test. The E R M S indicates that computational variations of standardized drought indices lead to highly variable results. From the LMM, the choice of the distribution and the observational window are the most important sources of uncertainty. They, on average, control between 19% and 63% (choice of distribution) and 24% to 70% (observation period) of the total variance of the SPI across all stations and month of the year, with similar values observed for the SPEI. The parameter estimation method and the GOF-tests, however, have almost no effect on the standardized indices. Total errors and observation period uncertainty are typically decreasing with the record length as one would expect, while the distribution uncertainty is almost independent from the record length. An additional assessment shows that the uncertainties are similar at the pan-European scale leading to uncertain characterizations of major events such as the drought of 2015. Overall, the uncertainty of standardized drought indices is substantial. Alternative approaches as nonparametric methods, ensemble approaches or probability-based indices based on established methods of extreme-value statistics should be considered, to make the indices more accurate. • A novel error model sheds light on the accuracy of drought indices SPI and SPEI. • Main sources of uncertainty are observation period and choice of distribution. • Parameter estimation methods and GOF tests have almost no effect. • Errors are substantial and may yield to false classifications of drought events. • Concepts are discussed to make drought indices more accurate.

Estimation and statistical inferences of variance components in the analysis of single-case experimental design using multilevel modeling.

Multilevel models (MLMs) can be used to examine treatment heterogeneity in single-case experimental designs (SCEDs). With small sample sizes, common issues for estimating between-case variance components in MLMs include nonpositive definite matrix, biased estimates, misspecification of covariance structures, and invalid Wald tests for variance components with bounded distributions. To address these issues, unconstrained optimization, model selection procedure based on parametric bootstrap, and restricted likelihood ratio test (RLRT)-based procedure are introduced. Using simulation studies, we compared the performance of two types of optimization methods (constrained vs. unconstrained) when the covariance structures are correctly specified or misspecified. We also examined the performance of a model selection procedure to obtain the optimal covariance structure. The results showed that the unconstrained optimization can avoid nonpositive definite issues to a great extent without a compromise in model convergence. The misspecification of covariance structures would cause biased estimates, especially with small between case variance components. However, the model selection procedure was found to attenuate the magnitude of bias. A practical guideline was generated for empirical researchers in SCEDs, providing conditions under which trustworthy point and interval estimates can be obtained for between-case variance components in MLMs, as well as the conditions under which the RLRT-based procedure can produce acceptable empirical type I error rate and power.

An Alternative Procedure to Produce a P-Spline Small Area Estimation Model Based on Partial Residual Plot and Significance Test of Spline Term

This study provides an alternative procedure to produce the penalized spline small area estimation (P-Spline SAE) model by plotting each of auxiliary variable and variable interest as a simple nonlinearity identification and performing the iterative procedure: by estimating the model, producing partial residual plots to check model adequacy and also identify the nonlinearity, and testing the significance of spline term using Restricted Likelihood Ratio Test (RLRT). These procedures are applied to estimate the monthly average per-capita expenditure of district level in the Province of Bali, 2014 using direct survey estimates from the National Socio-Economic Survey (Susenas 2014) and auxiliary variables from the administrative record of village data (PODES 2014). The Fay-Herriot (FH) model (M1) as a benchmark and four P-Spline SAE model, i.e. the P-Spline SAE with spline term: x2 and x3 (M2), x3 and x4 (M3), x2, x3 and x4 (M4), and x1, x2, x3 and x4 (M5), are obtained. From RLRT, the spline term of M3, M4, and M5 are statistically significant. According to the parametric bootstrap of mean squared prediction error (MSPE) and coefficient of variation (CV), the M4 and M5 show a significant improvement with the CV values range from about 1%-6% compared to M1 with a range from about 4%-17%, shows that these two models more efficient. The M3 model shows the opposite performance even though has the smallest AIC value. More detail, the MSPE and CV produced by M4 are slightly better than M5 makes the M4 is the best model in this study.

On the finite sample distribution of the likelihood ratio statistic for testing heterogeneity in meta-analysis.

In meta-analysis, hypothesis testing is one of the commonly used approaches for assessing whether heterogeneity exists in effects between studies. The literature concluded that the Q-statistic is clearly the best choice and criticized the performance of the likelihood ratio test in terms of the type I error control and power. However, all the criticism for the likelihood ratio test is based on the use of a mixture of two chi-square distributions with 0 and 1 degrees of freedom, which is justified only asymptotically. In this study, we develop a novel method to derive the finite sample distribution of the likelihood ratio test and restricted likelihood ratio test statistics for testing the zero variance component in the random effects model for meta-analysis. We also extend this result to the heterogeneity test when metaregression is applied. A numerical study shows that the proposed statistics have superior performance to the Q-statistic, especially when the number of studies collected for meta-analysis is small tomoderate.

Generalized linear mixed models with Gaussian mixture random effects: Inference and application

We propose a new class of generalized linear mixed models with Gaussian mixture random effects for clustered data. To overcome the weak identifiability issues, we fit the model using a penalized Expectation Maximization (EM) algorithm, and develop sequential locally restricted likelihood ratio tests to determine the number of components in the Gaussian mixture. Our work is motivated by an application to nationwide kidney transplant center evaluation in the United States, where the patient-level post-surgery outcomes are repeated measures of the care quality of the transplant centers. By taking into account patient-level risk factors and modeling the center effects by a finite Gaussian mixture model, the proposed model provides a convenient framework to study the heterogeneity among the transplant centers and controls the false discovery rate when screening for transplant centers with non-standard performance.

Exact variance component tests for longitudinal microbiome studies

In metagenomic studies, testing the association between microbiome composition and clinical outcomes translates to testing the nullity of variance components. Motivated by a lunghuman immunodeficiency virus (HIV) microbiome project, we study longitudinal microbiome data by using variance component models with more than two variance components. Current testing strategies only apply to models with exactly two variance components and when sample sizes are large. Therefore, they are not applicable to longitudinal microbiome studies. In this paper, we propose exact tests (score test, likelihood ratio test, and restricted likelihood ratio test) to (a) test the association of the overall microbiome composition in a longitudinal design and(b) detect the association of one specific microbiome cluster while adjusting for the effects from related clusters. Our approach combines the exact tests for null hypothesis with a single variance component with a strategy of reducing multiple variance components to a single one. Simulation studies demonstrate that our method has a correct type I error rate and superior power compared to existing methods at small sample sizes and weak signals. Finally, we apply our method to a longitudinal pulmonary microbiome study ofHIV-infected patients and reveal two interesting genera Prevotella and Veillonella associated with forced vital capacity. Our findings shed light on the impact of the lung microbiome on HIV complexities. The method is implemented in the open-source, high-performance computing language Julia and is freely available at https://github.com/JingZhai63/VCmicrobiome.

Composite kernel machine regression based on likelihood ratio test for joint testing of genetic and gene-environment interaction effect.

Most common human diseases are a result from the combined effect of genes, the environmental factors, and their interactions such that including gene-environment (GE) interactions can improve power in gene mapping studies. The standard strategy is to test the SNPs, one-by-one, using a regression model that includes both the SNP effect and the GE interaction. However, the SNP-by-SNP approach has serious limitations, such as the inability to model epistatic SNP effects, biased estimation, and reduced power. Thus, in this article, we develop a kernel machine regression framework to model the overall genetic effect of a SNP-set, considering the possible GE interaction. Specifically, we use a composite kernel to specify the overall genetic effect via a nonparametric function andwe model additional covariates parametrically within the regression framework. The composite kernel is constructed as a weighted average of two kernels, one corresponding to the genetic main effect and one corresponding to the GE interaction effect. We propose a likelihood ratio test (LRT) and a restricted likelihood ratio test (RLRT) for statistical significance. We derive a Monte Carlo approach for the finite sample distributions of LRT and RLRT statistics. Extensive simulations and real data analysis show that our proposed method has correct type I error and can have higher power than score-based approaches under many situations.

Fiducial generalized p-values for testing zero-variance components in linear mixed-effects models

Linear mixed-effects models are widely used in analysis of longitudinal data. However, testing for zero-variance components of random effects has not been well-resolved in statistical literature, although some likelihood-based procedures have been proposed and studied. In this article, we propose a generalized $p$-value based method in coupling with fiducial inference to tackle this problem. The proposed method is also applied to test linearity of the nonparametric functions in additive models. We provide theoretical justifications and develop an implementation algorithm for the proposed method. We evaluate its finite-sample performance and compare it with that of the restricted likelihood ratio test via simulation experiments. We illustrate the proposed approach using an application from a nutritional study.

Boosting Gene Mapping Power and Efficiency with Efficient Exact Variance Component Tests of Single Nucleotide Polymorphism Sets.

Single nucleotide polymorphism (SNP) set tests have been a powerful method in analyzing next-generation sequencing (NGS) data. The popular sequence kernel association test (SKAT) method tests a set of variants as random effects in the linear mixed model setting. Its P-value is calculated based on asymptotic theory that requires a large sample size. Therefore, it is known that SKAT is conservative and can lose power at small or moderate sample sizes. Given the current cost of sequencing technology, scales of NGS are still limited. In this report, we derive and implement computationally efficient, exact (nonasymptotic) score (eScore), likelihood ratio (eLRT), and restricted likelihood ratio (eRLRT) tests, ExactVCTest, that can achieve high power even when sample sizes are small. We perform simulation studies under various genetic scenarios. Our ExactVCTest (i.e., eScore, eLRT, eRLRT) exhibits well-controlled type I error. Under the alternative model, eScore P-values are universally smaller than those from SKAT. eLRT and eRLRT demonstrate significantly higher power than eScore, SKAT, and SKAT optimal (SKAT-o) across all scenarios and various samples sizes. We applied these tests to an exome sequencing study. Our findings replicate previous results and shed light on rare variant effects within genes. The software package is implemented in the open source, high-performance technical computing language Julia, and is freely available at https://github.com/Tao-Hu/VarianceComponentTest.jl Analysis of each trait in the exome sequencing data set with 399 individuals and 16,619 genes takes around 1 min on a desktop computer.

Testing for additivity in non‐parametric regression

AbstractThis article discusses a novel approach for testing for additivity in non‐parametric regression. We represent the model using a linear mixed model framework and equivalently rewrite the original testing problem as testing for a subset of zero variance components. We propose two testing procedures: the restricted likelihood ratio test and the generalized F test. We develop the finite sample null distribution of the restricted likelihood ratio test and generalized F test using the spectral decomposition of the restricted likelihood ratio and the residual sum of squares, respectively. The null distribution is non‐standard and we provide a fast algorithm to simulate from the null distribution of the tests. We show, through numerical investigation, that the proposed testing procedures outperform the available alternatives and apply the methods to a diabetes data set. The Canadian Journal of Statistics 44: 445–462; 2016 © 2016 Statistical Society of Canada

<b>Mixed models in cerebral ischemia study

The data modeling from longitudinal studies stands out in the current scientific scenario, especially in the areas of health and biological sciences, which induces a correlation between measurements for the same observed unit. Thus, the modeling of the intra-individual dependency is required through the choice of a covariance structure that is able to receive and accommodate the sample variability. However, the lack of methodology for correlated data analysis may result in an increased occurrence of type I or type II errors and underestimate/overestimate the standard errors of the model estimates. In the present study, a Gaussian mixed model was adopted for the variable response latency of an experiment investigating the memory deficits in animals subjected to cerebral ischemia when treated with fish oil (FO). The model parameters estimation was based on maximum likelihood methods. Based on the restricted likelihood ratio test and information criteria, the autoregressive covariance matrix was adopted for errors. The diagnostic analyses for the model were satisfactory, since basic assumptions and results obtained corroborate with biological evidence; that is, the effectiveness of the FO treatment to alleviate the cognitive effects caused by cerebral ischemia was found.

Restricted Likelihood Ratio Tests in Predictive Regression

Chen and Deo (2009a) proposed procedures based on restricted maximum likelihood (REML) for estimation and inference in the context of predictive regression. Their method achieves bias reduction in both estimation and inference which assists in overcoming size distortion in predictive hypothesis testing. This paper provides extensions of the REML approach to more general cases which allow for drift in the predictive regressor and multiple regressors. It is shown that without modification the REML approach is seriously oversized and can have unit rejection probability in the limit under the null when the drift in the regressor is dominant. A limit theory for the modified REML test is given under a localized drift specification that accommodates predictors with varying degrees of persistence. The extension is useful in empirical work where predictors typically involve stochastic trends with drift and where there are multiple regressors. Simulations show that with these modifications, the good performance of the restricted likelihood ratio test (RLRT) is preserved and that RLRT outperforms other predictive tests in terms of size and power even when there is no drift in the regressor.

Restricted Likelihood Ratio Tests for Functional Effects in the Functional Linear Model

The goal of our article is to provide a transparent, robust, and computationally feasible statistical approach for testing in the context of scalar-on-function linear regression models. Assuming linearity between response and predictors, we are interested in testing for the necessity of functional effects. Our methods are motivated by and applied to a large longitudinal study involving diffusion tensor imaging of intracranial white matter tracts in a susceptible cohort. In the context of this study, we conduct hypothesis tests that are motivated by anatomical knowledge and support recent findings regarding the relationship between cognitive impairment and white matter demyelination. R code and data are in the examples of refund::rlrt.pfr(). Supplementary materials for this article are available online.

Likelihood ratio tests in rare variant detection for continuous phenotypes.

It is believed that rare variants play an important role in human phenotypes; however, the detection of rare variants is extremely challenging due to their very low minor allele frequency. In this paper, the likelihood ratio test (LRT) and restricted likelihood ratio test (ReLRT) are proposed to test the association of rare variants based on the linear mixed effects model, where a group of rare variants are treated as random effects. Like the sequence kernel association test (SKAT), a state-of-the-art method for rare variant detection, LRT and ReLRT can effectively overcome the problem of directionality of effect inherent in the burden test in practice. By taking full advantage of the spectral decomposition, exact finite sample null distributions for LRT and ReLRT are obtained by simulation. We perform extensive numerical studies to evaluate the performance of LRT and ReLRT, and compare to the burden test, SKAT and SKAT-O. The simulations have shown that LRT and ReLRT can correctly control the type I error, and the controls are robust to the weights chosen and the number of rare variants under study. LRT and ReLRT behave similarly to the burden test when all the causal rare variants share the same direction of effect, and outperform SKAT across various situations. When both positive and negative effects exist, LRT and ReLRT suffer from few power reductions compared to the other two competing methods; under this case, an additional finding from our simulations is that SKAT-O is no longer the optimal test, and its power is even lower than that of SKAT. The exome sequencing SNP data from Genetic Analysis Workshop 17 were employed to illustrate the proposed methods, and interesting results are described.

Restricted likelihood ratio tests for linearity in scalar-on-function regression

We propose a procedure for testing the linearity of a scalar-on-function regression relationship. To do so, we use the functional generalized additive model (FGAM), a recently developed extension of the functional linear model. For a functional covariate X(t), the FGAM models the mean response as the integral with respect to t of F{X(t),t} where F is an unknown bivariate function. The FGAM can be viewed as the natural functional extension of generalized additive models. We show how the functional linear model can be represented as a simple mixed model nested within the FGAM. Using this representation, we then consider restricted likelihood ratio tests for zero variance components in mixed models to test the null hypothesis that the functional linear model holds. The methods are general and can also be applied to testing for interactions in a multivariate additive model or for testing for no effect in the functional linear model. The performance of the proposed tests is assessed on simulated data and in an application to measuring diesel truck emissions, where strong evidence of nonlinearities in the relationship between the functional predictor and the response are found.

Rare Variants Detection with Kernel Machine Learning Based on Likelihood Ratio Test

This paper mainly utilizes likelihood-based tests to detect rare variants associated with a continuous phenotype under the framework of kernel machine learning. Both the likelihood ratio test (LRT) and the restricted likelihood ratio test (ReLRT) are investigated. The relationship between the kernel machine learning and the mixed effects model is discussed. By using the eigenvalue representation of LRT and ReLRT, their exact finite sample distributions are obtained in a simulation manner. Numerical studies are performed to evaluate the performance of the proposed approaches under the contexts of standard mixed effects model and kernel machine learning. The results have shown that the LRT and ReLRT can control the type I error correctly at the given α level. The LRT and ReLRT consistently outperform the SKAT, regardless of the sample size and the proportion of the negative causal rare variants, and suffer from fewer power reductions compared to the SKAT when both positive and negative effects of rare variants are present. The LRT and ReLRT performed under the context of kernel machine learning have slightly higher powers than those performed under the context of standard mixed effects model. We use the Genetic Analysis Workshop 17 exome sequencing SNP data as an illustrative example. Some interesting results are observed from the analysis. Finally, we give the discussion.

Massively Parallel Nonparametric Regression, With an Application to Developmental Brain Mapping

A penalized approach is proposed for performing large numbers of parallel nonparametric analyses of either of two types: restricted likelihood ratio tests of a parametric regression model versus a general smooth alternative, and nonparametric regression. Compared with naïvely performing each analysis in turn, our techniques reduce computation time dramatically. Viewing the large collection of scatterplot smooths produced by our methods as functional data, we develop a clustering approach to summarize and visualize these results. Our approach is applicable to ultra-high-dimensional data, particularly data acquired by neuroimaging; we illustrate it with an analysis of developmental trajectories of functional connectivity at each of approximately 70,000 brain locations. Supplementary materials, including an appendix and an R package, are available online.

Linear Score Tests for Variance Components in Linear Mixed Models and Applications to Genetic Association Studies

Following the rapid development of genome-scale genotyping technologies, genetic association mapping has become a popular tool to detect genomic regions responsible for certain (disease) phenotypes, especially in early-phase pharmacogenomic studies with limited sample size. In response to such applications, a good association test needs to be (1)applicable to a wide range of possible genetic models, including, but not limited to, the presence of gene-by-environment or gene-by-gene interactions and non-linearity of a group of marker effects, (2)accurate in small samples, fast to compute on the genomic scale, and amenable to large scale multiple testing corrections, and (3)reasonably powerful to locate causal genomic regions. The kernel machine method represented in linear mixed models provides a viable solution by transforming the problem into testing the nullity of variance components. In this study, we consider score-based tests by choosing a statistic linear in the score function. When the model under the null hypothesis has only one error variance parameter, our test is exact in finite samples. When the null model has more than one variance parameter, we develop a new moment-based approximation that performs well in simulations. Through simulations and analysis of real data, we demonstrate that the new test possesses most of the aforementioned characteristics, especially when compared to existing quadratic score tests or restricted likelihood ratio tests.