Composite Likelihood Research Articles

Exploiting multivariate network meta-analysis: A calibrated Bayesian composite likelihood inference.

Multivariate network meta-analysis has emerged as a powerful tool in evidence synthesis by incorporating multiple outcomes and treatments. Despite its advantages, this method comes with methodological challenges, such as the issue of unreported within-study correlations among treatments and outcomes, which potentially lead to misleading conclusions. In this paper, we proposed a calibrated Bayesian composite likelihood approach to overcome this limitation. The proposed method eliminated the need to specify a full likelihood function while allowing for the unavailability of within-study correlations among treatments and outcomes. Additionally, we developed a hybrid Gibbs sampler algorithm along with the Open-Faced Sandwich post-sampling adjustment to enable robust posterior inference. Through comprehensive simulation studies, we demonstrated that the proposed approach yielded unbiased estimates while maintaining coverage probabilities close to the nominal level. Furthermore, we implemented the proposed method on two real-world network meta-analysis datasets; one comparing treatment procedures for the root coverage and another comparing treatments for anaemia in chronic kidney disease patients.

Network meta-analysis made simple: a composite likelihood approach.

What is already known?: Network meta-analysis extends conventional pairwise meta-analysis by simultaneously synthesizing multiple interventions in a single integrated analysis.A significant challenge in network meta-analysis is the lack of reported within-study correlations among treatment comparisons in the published studies.What is new?: We propose a new method for network meta-analysis that ensures vaild statistical inference without the need for knowledge of within-study correlations.The proposed method employs a composite likelihood and a sandwich-type robust variance estimator, offering a computationally efficient and scalable solution, particularly for network meta-analysis with a large number of treatments and studies.Potential impact for Research Synthesis Methods readers: The proposed method can be easily applied to any univariate network meta-analysis project without requiring knowledge of within-study correlations among treatment comparisons.

An efficient workflow for modelling high-dimensional spatial extremes

We develop a comprehensive methodological workflow for Bayesian modelling of high-dimensional spatial extremes that lets us describe both weakening extremal dependence at increasing levels and changes in the type of extremal dependence class as a function of the distance between locations. This is achieved with a latent Gaussian version of the spatial conditional extremes model that allows for computationally efficient inference with R-INLA. Inference is made more robust using a post hoc adjustment method that accounts for possible model misspecification. This added robustness makes it possible to extract more information from the available data during inference using a composite likelihood. The developed methodology is applied to the modelling of extreme hourly precipitation from high-resolution radar data in Norway. Inference is performed quickly, and the resulting model fit successfully captures the main trends in the extremal dependence structure of the data. The post hoc adjustment is found to further improve model performance.

Robust analysis of stepped wedge trials using composite likelihood models.

Stepped wedge trials (SWTs) are a type of cluster randomized trial that involve repeated measures on clusters and design-induced confounding between time and treatment. Although mixed models are commonly used to analyze SWTs, they are susceptible to misspecification particularly for cluster-longitudinal designs such as SWTs. Mixed model estimation leverages both "horizontal" or within-cluster information and "vertical" or between-cluster information. To use horizontal information in a mixed model, both the mean model and correlation structure must be correctly specified or accounted for, since time is confounded with treatment and measurements are likely correlated within clusters. Alternative non-parametric methods have been proposed that use only vertical information; these are more robust because between-cluster comparisons in a SWT preserve randomization, but these non-parametric methods are not very efficient. We propose a composite likelihood method that focuses on vertical information, but has the flexibility to recover efficiency by using additional horizontal information. We compare the properties and performance of various methods, using simulations based on COVID-19 data and a demonstration of application to the LIRE trial. We found that a vertical composite likelihood model that leverages baseline data is more robust than traditional methods, and more efficient than methods that use only vertical information. We hope that these results demonstrate the potential value of model-based vertical methods for SWTs with a large number of clusters, and that these new tools are useful to researchers who are concerned about misspecification of traditional models.

Whole-genome resequencing reveals diversity and selective signals in the Wuxue goat.

Animal genetic resources are crucial for ensuring global food security. However, in recent years, a noticeable decline in the genetic diversity of livestock has occurred worldwide. This decline is pronounced in developing countries, where the management of these resources is insufficient. In the current study, we performed whole genome sequencing for 20 Wuxue (WX) and five Guizhou White (GW) goats. Additionally, we utilized the published genomes of 131 samples representing five different goat breeds from various regions in China. We investigated and compared the genetic diversity and selection signatures of WX goats. Whole genome sequencing analysis of the WX and GW populations yielded 120 425 063 SNPs, which resided primarily in intergenic and intron regions. Population genetic structure revealed that WX exhibited genetic resemblance to GW, Chengdu Brown, and Jintang Black and significant differentiation from the other goat breeds. In addition, three methods (nucleotide diversity, linkage disequilibrium decay, and runs of homozygosity) showed moderate genetic diversity in WX goats. We used nucleotide diversity and composite likelihood ratio methods to identify within-breed signatures of positive selection in WX goats. A total of 369 genes were identified using both detection methods, including genes related to reproduction (GRID2, ZNF276, TCF25, and SPIRE2), growth (HMGA2 and GJA3), and immunity (IRF3 and SRSF3). Overall, this study explored the adaptability of WX goats, shedding light on their genetic richness and potential to thrive in challenges posed by climatic changes and diseases. Further investigations are warranted to harness these insights to enhance more efficient and sustainable goat breeding initiatives.

The maximally selected likelihood ratio test in random coefficient models

Summary In a recent contribution, we developed a family of cumulative sum-based change-point tests in the context of a random coefficient autoregressive model of order 1. In the current paper, we complement the results in that contribution by studying the (maximally selected) likelihood ratio statistic. We show that this has power versus breaks occurring even as close as periods from the beginning/end of sample; moreover, the use of quasi-maximum likelihood-based estimates yields better power properties, with the added bonus of being nuisance-free. Our test statistic has the same distribution—of the Darling–Erdős type—irrespective of whether the data are stationary or not, and can therefore be applied with no prior knowledge of this. Our simulations show that our test has very good power and, when applying a suitable correction to the asymptotic critical values, the correct size. We illustrate the usefulness and generality of our approach through applications to economic and epidemiological time series.

Fractional stochastic volatility model

This article introduces a discrete‐time fractional stochastic volatility model (FSV) based on fractional Gaussian noise. The new model includes the standard stochastic volatility model as a special case and has the same limit as the fractional integrated stochastic volatility (FISV) model, which is the continuous‐time fractional Ornstein–Uhlenbeck process. A simulated maximum likelihood method, which maximizes the time‐domain log‐likelihood function calculated by the importance sampling technique, and a frequency‐domain quasi maximum likelihood method (or quasi Whittle) are employed to estimate the model parameters. Simulation studies suggest that, while both estimation methods can accurately estimate the model, the simulated maximum likelihood method outperforms the quasi Whittle method. As an illustration, we fit the FSV and FISV models with the proposed estimation techniques to the S&P 500 composite index over a sample period spanning 45 years.

Comparative whole-genome resequencing to uncover selection signatures linked to litter size in Hu Sheep and five other breeds

Hu sheep (HS), a breed of sheep carrying the FecB mutation gene, is known for its “year-round estrus and multiple births” and is an ideal model for studying the high fecundity mechanisms of livestock. Through analyzing and comparing the genomic selection features of Hu sheep and other sheep breeds, we identified a series of candidate genes that may play a role in Hu sheep’s high fecundity mechanisms. In this study, we conducted whole-genome resequencing on six breeds and screened key mutations significantly correlated with high reproductive traits in sheep. Notably, the CC2D1B gene was selected by the fixation index (FST) and the cross-population composite likelihood ratio (XP-CLR) methods in HS and other five breeds. It was worth noting that the CC2D1B gene in HS was different from that in other sheep breeds, and seven missense mutations have been identified. Furthermore, the linkage disequilibrium (LD) analysis revealed a strong linkage disequilibrium in this specific gene region. Subsequently, by performing different grouping based on FecB genotypes in Hu sheep, genome-wide selective signal analysis screened several genes related to reproduction, such as BMPR1B and PPM1K. Besides, FST analysis identified functional genes related to reproductive traits, including RHEB, HSPA2, PPP1CC, HVCN1, and CCDC63. Additionally, a missense mutation was found in the CCDC63 gene and the haplotype was different between the high reproduction (HR) group and low reproduction (LR) group in HS. In summary, we discovered genetic differentiation among six distinct breeding sheep breeds at the whole genome level. Additionally, we identified a set of genes which were associated with reproductive performance in Hu sheep and visualized how these genes differed in different breeds. These findings laid a theoretical foundation for understanding genetic mechanisms behind high prolific traits in sheep.

Threshold Network GARCH Model

Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model and its variations have been widely adopted in the study of financial volatilities, while the extension of GARCH‐type models to high‐dimensional data is always difficult because of over‐parameterization and computational complexity. In this article, we propose a multi‐variate GARCH‐type model that can simplify the parameterization by utilizing the network structure that can be appropriately specified for certain types of high‐dimensional data. The asymmetry in the dynamics of volatilities is also considered as our model adopts a threshold structure. To enable our model to handle data with extremely high dimension, we investigate the near‐epoch dependence (NED) of our model, and the asymptotic properties of our quasi‐maximum‐likelihood‐estimator (QMLE) are derived from the limit theorems for NED random fields. Simulations are conducted to test our theoretical results. At last we fit our model to log‐returns of four groups of stocks and the results indicate that bad news is not necessarily more influential on volatility if the network effects are considered.

Statistical Modeling of Right-Censored Spatial Data Using Gaussian Random Fields

Consider a fixed number of clustered areas identified by their geographical coordinates that are monitored for the occurrences of an event such as a pandemic, epidemic, or migration. Data collected on units at all areas include covariates and environmental factors. We apply a probit transformation to the time to event and embed an isotropic spatial correlation function into our models for better modeling as compared to existing methodologies that use frailty or copula. Composite likelihood technique is employed for the construction of a multivariate Gaussian random field that preserves the spatial correlation function. The data are analyzed using counting process and geostatistical formulation that led to a class of weighted pairwise semiparametric estimating functions. The estimators of model parameters are shown to be consistent and asymptotically normally distributed under infill-type asymptotic spatial statistics. Detailed small sample numerical studies that are in agreement with theoretical results are provided. The foregoing procedures are applied to the leukemia survival data in Northeast England. A comparison to existing methodologies provides improvement.

Non-adaptive estimation for degenerate diffusion processes

We consider a degenerate system of stochastic differential equations. The first component of the system has a parameter θ 1 \theta _1 in a non-degenerate diffusion coefficient and a parameter θ 2 \theta _2 in the drift term. The second component has a drift term with a parameter θ 3 \theta _3 and no diffusion term. Parametric estimation of the degenerate diffusion system is discussed under a sampling scheme. We investigate the asymptotic behavior of the joint quasi-maximum likelihood estimator for ( θ 1 , θ 2 , θ 3 ) (\theta _1,\theta _2,\theta _3) . The estimation scheme is non-adaptive. The estimator incorporates information of the increments of both components, and under this construction, we show that the asymptotic variance of the estimator for θ 1 \theta _1 is smaller than the one for standard estimator based on the first component only, and that the convergence of the estimator for θ 3 \theta _3 is much faster than for the other parameters. By simulation studies, we compare the performance of the joint quasi-maximum likelihood estimator with the adaptive and one-step estimators investigated in Gloter and Yoshida [Electron. J. Statist 15 (2021), no. 1, 1424–1472].

Inference of Dynamic Spatial Panel Data Model and Its Application in Carbon Emission Analysis

The Spatial Dynamic Panel Data (SDPD) model is a widely used statistical model in the fields of economics and social sciences and has been the subject of extensive research by many scholars in recent years. Existing methods for parameter estimation primarily focus on improvements to the Generalized Method of Moments (GMM) and Quasi-Maximum Likelihood (QML). In this paper, we employ the method of Empirical Likelihood (EL) for statistical inference of the dynamic spatial panel data model and obtain confidence regions for the parameters. Through numerical simulations, we present the performance of the confidence regions obtained using the Empirical Likelihood (EL) and the Asymptotically Normal (NA) methods under finite samples and compare the two approaches. Finally, we analyze carbon using the suggested model and techniques.

Complete mitochondrial genome of critically endangered catfish Hemibagrus punctatus (Jerdon, 1849) and comparative analysis for insights into the phylogeny of hemibagrids through mitogenomic approach.

Hemibagrus punctatus (Jerdon, 1849) is a critically endangered bagrid catfish endemic to the Western Ghats of India, whose population is declining due to anthropogenic activities. The current study aims to compare the mitogenome of H. punctatus with that of other Bagrid catfishes and provide insights into their evolutionary relationships. Samples were collected from Hemmige Karnataka, India. In the present study, the mitogenome of H. punctatus was successfully assembled, and its phylogenetic relationships with other Bagridae species were studied. The total genomic DNA of samples was extracted following the phenol-chloroform isoamyl alcohol method. Samples were sequenced, and the Illumina paired-end reads were assembled to a contig length of 16,517bp. The mitochondrial genome was annotated using MitoFish and MitoAnnotator (Iwasaki et al., 2013). A robust phylogenetic analysis employing NJ (Maximum composite likelihood) and ASAP methods supports the classification of H. punctatus within the Bagridae family, which validates the taxonomic status of this species. In conclusion, this research enriches our understanding of H. punctatus mitogenome, shedding light on its evolutionary dynamics within the Bagridae family and contributing to the broader knowledge of mitochondrial genes in the context of evolutionary biology. The study's findings contribute to a better understanding of the mitogenome of H. punctatus and provide insights into the evolutionary relationships within other Hemibagrids.

Molecular Characterization and Screening of Protease Production from indigenous thermophilic fungi Malbranchea cinnamomea isolated from dairy farm compost soil

Compost environments represent a rich reservoir of microbial diversity, including thermophilic fungi with the potential to produce proteases. The dynamic and nutrient-rich nature of compost provides a conducive habitat for the growth and metabolism of diverse fungal species, including Malbranchea cinnamomea. In this study, Malbranchea cinnamomea was isolated from dairy farm compost soil and characterized based on its morphological characters and subsequently confirmed molecularly using internal transcribed spacer (ITS) sequencing. The Phylogenetic tree was inferred using Neighbour- joining method. The evolutionary distances were computed using the maximum composite likelihood method. The identified fungi was primarily screened for protease and further used for enzyme production in submerged fermentation for 4-6 days at 50OC. Culture filtrate obtained were used for the assessment of enzymatic activity. This study revealed thermophilic fungus Malbranchea cinnamomea holds significant potential for protease production. The enzyme activity was estimated at 50 °C, pH 7, six days incubation time and using 1.0 ml inoculum. Under submerged substrate cultivation, the best substrate lodge for maximal fungal biomass production was YPSs broth. Hence this isolate could be further exploited for numerous industrial applications that require thermophilic enzymes.

On efficiency of locally D -optimal designs under heteroscedasticity and non-Gaussianity

On efficiency of locally <i>D</i> -optimal designs under heteroscedasticity and non-Gaussianity

A dynamic count process

The current paper aims to complement the recent development of the observation-driven models of dynamic counts with a parametric-driven one for a general case, particularly discrete two parameters exponential family distributions. The current paper proposes a finite semiparametric exponential mixture of SETAR processes of the conditional mean of counts to capture the nonlinearity and complexity. Because of the intrinsic latency of the conditional mean, the general additive state-space representation of dynamic counts is firstly proposed then stationarity and geometric ergodicity are established under a mild set of conditions. We also propose to estimate the unknown parameters by using quasi maximum likelihood estimation and establishes the asymptotic properties of the quasi maximum likelihood estimators (QMLEs), particularly T-consistency and normality under the relatively mild set of conditions. Furthermore, the finite sample properties of the QMLEs are investigated via simulation exercises and an illustration of the proposed process is presented by applying the proposed method to the intraday transaction counts per minute of AstraZeneca stock.

Quasi-maximum likelihood estimation and penalized estimation under non-standard conditions

Quasi-maximum likelihood estimation and penalized estimation under non-standard conditions

Varying coefficient panel data models and methods under correlated error components: Application to disparities in mental health services in England

The contribution of this paper is twofold. Firstly, it introduces novel regression models that combine two important areas of the methodological development in panel data analysis, namely a varying coefficient specification and spatial error dependence. The former allows relatively flexible nonlinear interactions; the latter enables spatial correlations of the disturbance and thus differ significantly from the other random effect models in the literature. To estimate the model, a new estimation procedure is established that can be viewed as a generalization of the quasi-maximum likelihood method for a spatial panel data model to the well-known conditional local likelihood procedure. Novel inference methods, particularly variable selection and hypothesis testing of the parameter constancy, are introduced and are shown to be effective under the complex spatial error dependence. Equally importantly, this paper makes a substantial contribution to the understanding of financing and expenditure for health and social care. In particular, we empirically analyze and explain the effects of political ideologies on the local fiscal policy in England, especially the expenditure on mental health services.

Nonlinear Spatial Dynamic Panel Data Models with Endogenous Dominant Units: An Application to Share Data

This article develops a nonlinear spatial dynamic panel data model with one particularly interesting application to a structural interaction model for share data. To account for effects from dominant (popular) units, the spatial weights matrix in our model can allow for unbounded column sums. To account for heterogeneity, our model includes two-way fixed effects and heteroscedastic errors. We further consider the potential endogeneity of the spatial weight matrix constructed from socioeconomic distance. We investigate the quasi-maximum likelihood estimator (QMLE), generalized methods of moments estimator (GMME), and root estimator (RTE), and establish their consistency and asymptotic normality based on the near epoch dependence (NED) framework. The RTE can derive a relatively computationally simple and closed-form solution without evaluating the QMLE’s Jacobian matrix as well as the iterations by GMME. We consider both n , T → ∞ , and the strength of the dominant units is equal to 1 when T → ∞ . For the purpose of empirical analysis, we derive the marginal effects and their limiting distributions based on the proposed estimators. In an empirical application, we apply our model to China’s prefecture city-level data, revealing significant spillover effects of the tertiary industry share. These findings suggest that the development of the tertiary sector in large cities can foster its growth in small cities.

Whole genome resequencing reveals genomic regions related to red plumage in ducks

Plumage color is a characteristic trait of ducks that originates as a result of natural and artificial selection. As a conspicuous phenotypic feature, it is a breed characteristic. Previous studies have identified some genes associated with the formation of black and white plumage in ducks. However, studies on the genetic basis underlying the red plumage phenotype in ducks are limited. Here, genome-wide association analysis (GWAS) and selection signal detection (Fst, θπ ratio, and cross-population composite likelihood ratio [XP-CLR]) were conducted to identify candidate regions and genes underlying duck plumage color phenotype. Selection signal detection revealed 29 overlapping genes (including ENPP1 and ULK1) significantly associated with red plumage color in Ji'an Red ducks. ENSAPLG00000012679, ESRRG, and SPATA5 were identified as candidate genes associated with red plumage using GWAS. Selection signal detection revealed that 19 overlapping genes (including GMDS, PDIA6, and ODC1) significantly correlated with light brown plumage in Brown Tsaiya ducks. GWAS to narrow down the significant regions further revealed nine candidate genes (AKT1, ATP6V1C2, GMDS, LRP4, MAML3, PDIA6, PLD5, TMEM63B, and TSPAN8). Notably, in Brown Tsaiya ducks, GMDS, ODC1, and PDIA6 exhibit significantly differentiated allele frequencies among other feather-colored ducks, while in Ji'an Red ducks, ENSAPLG00000012679 has different allele frequency distributions compared with that in other feather-colored ducks. This study offers new insights into the variation and selection of the red plumage phenotype using GWAS and selective signals.