Likelihood Procedure Research Articles

Consistent non-Gaussian pseudo maximum likelihood estimators

We characterise the mean and variance parameters that distributionally misspecified maximum likelihood estimators can consistently estimate in location-scale models, and provide simple closed-form consistent estimators for the rest. Including means and a multivariate coverage make our procedures useful for Garch-M models and empirically relevant macro and finance applications involving Vars and multivariate regressions. We study the statistical properties of our proposed consistent estimators, as well as their efficiency relative to Gaussian pseudo maximum likelihood and semiparametric procedures. We provide finite sample results through Monte Carlo simulations. Finally, we discuss two practical applications to individual stock returns and mean–variance efficiency/spanning tests.

A data-space inversion procedure for well control optimization and closed-loop reservoir management

Data-space inversion (DSI) methods provide posterior (history-matched) predictions for quantities of interest, along with uncertainty quantification, without constructing posterior models. Rather, predictions are generated directly from a large set of prior model simulations and observed data. In this work, we develop a data-space inversion with variable control (DSIVC) procedure that enables forecasting with user-specified well controls in the post-history-match prediction period. In DSIVC, flow simulations on all prior realizations, with randomly sampled well controls, are first performed. User-specified controls are treated as additional observations to be matched in posterior predictions. Posterior data samples are generated using a randomized maximum likelihood procedure with a gradient-based optimizer. For prescribed post-history-match well control settings, posterior predictions can be generated in seconds or minutes. Results are presented for a channelized system, and posterior predictions from DSIVC are compared with those from the standard DSI method. Standard DSI requires prior models to be re-simulated using the specified controls, while DSIVC requires only one set of prior simulations. Substantial uncertainty reduction is achieved through data-space inversion, and reasonable agreement between DSIVC and DSI results is generally observed. DSIVC is then applied for data assimilation combined with production optimization under uncertainty, as well as for closed-loop reservoir management, which entails a sequence of data assimilation and optimization steps. Clear improvement in the objective function is attained in these examples.

Avantium to locate new plant-based monoethylene glycol plant in Delfzijl

The aim of this study was to investigate the genetic and phenotypic aspects of growth efficiency-related traits in Mazandaran native chickens. For this purpose, phenotypic and pedigree data of 21 generations were obtained from Mazandaran native chicken breeding center. Studied traits included Kleiber ratio (KR), relative growth rate (RGR) and growth efficiency (EF) at different growth phases from hatch to 8 weeks of age (1-8), hatch to 12 weeks of age (1-12), 8 weeks to 12 weeks of age (8-12), hatch to sex maturity (1-SM), 8 weeks of age to sex maturity (8-SM) and 12 weeks of age to sex maturity (12-SM). Genetic parameters and (Co) variance components were estimated by Restricted Maximum Likelihood (REML) procedure under six different univariate animal models using WOMBAT software with and without considering the maternal effects. Based on the results of the best model, determined by the Akaike Information Criterion (AIC), the estimates of direct heritability ranged from 0.05 (EF8-12) to 0.32 (EF12-SM) for growth efficiency, from 0.06 (KR8-12) to 0.37 (KR8 -SM) for Kleiber ratios and from 0.06 (RGR8-12) to 0.16 (RGR1-12) for relative growth rates. Maternal heritability ranged from 0.03 (EF8-12, EF1-8) to 0.08 (EF1-SM) for growth efficiency, from 0.006 (KR8-12) to 0.097 (KR1-SM) for Kleiber ratios and form 0.03 (RGR8-12) to 0.14 (RGR1-12) for relative growth rates. Negative genetic correlations between direct and maternal additive genetic effects were estimated for all studied traits. In conclusion, significant maternal effects indicated that these effects should be considered in genetic evaluation of traits.

Genetic parameters for hair characteristics and core body temperature in a multibreed Brahman-Angus herd1.

Thermal stress in hot humid conditions limits cattle production. The objectives for this study were to estimate genetic parameters for hair characteristics and core body temperature under low and high temperature humidity index (THI) conditions. Hair samples were collected and measured for length and diameter. Core body temperature was measured as vaginal temperature every 15 min over a 5-d period using an iButton temperature measuring device implanted in a blank CIDR in 336 heifers from the University of Florida multibreed herd (ranging from 100% Angus to 100% Brahman). Restricted maximum likelihood procedures were used to estimate heritabilities from multiple bivariate animal models using the WOMBAT program. Estimates of heritability for hair diameter, undercoat length, topcoat length, body temperature under low THI conditions, and body temperature under high THI conditions were 0.50, 0.67, 0.42, 0.32, and 0.26, respectively. The genetic parameters estimated in this study indicate a large, exploitable genetic variance which can be selected upon to improve tolerance in cattle. Breed effects for differing compositions of Brahman and Angus were also estimated. As Brahman breed composition increased by 25% undercoat length, topcoat length, body temperature under low THI conditions, and body temperature under high THI conditions decreased by 1.32 mm, 2.94 mm, 0.11 °C, and 0.14 °C, respectively. Under both low and high THI conditions, cattle with 25% Brahman breed composition or greater maintained a significantly lower body temperature than the 100% Angus breed group. The incorporation of Brahman germplasm is recommended for herds that often experience heat stress conditions in order to increase resilience to heat stress.

Residual and influence analysis to a general class of simplex regression

In this paper, we propose a residual and local influence analysis for diagnostics in a general class of simplex regression model. Here, we introduce this class in which the predictors involve covariates and nonlinear functions in the parameters. We provide closed-form expressions for the score functions, information matrices, as well a procedure for the choice of initial guesses to be used in the Fisher’s iterative scheme for the estimation by maximum likelihood. All diagnostic techniques were also adjusted for the linear simplex model. We present Monte Carlo simulations to investigate the empirical distribution of the proposed residual and the performance of the starting points scheme. We also performed three applications to real data; one of them explores the features of nonlinear regressions. By performing diagnostic analysis, we compare the beta and simplex fits to two datasets. The applications results favor the simplex regression to fit data close to the boundaries of the unit interval. Indeed, the simplex regression can present estimation by maximum likelihood procedure more robust to influential cases than the beta regression.

Local composite partial likelihood estimation for length-biased and right-censored data

ABSTRACTLength-biased data, which are often encountered in engineering, economics and epidemiology studies, are generally subject to right censoring caused by the research ending or the follow-up loss. The structure of length-biased data is distinct from conventional survival data, since the independent censoring assumption is often violated due to the biased sampling. In this paper, a proportional hazard model with varying coefficients is considered for the length-biased and right-censored data. A local composite likelihood procedure is put forward for the estimation of unknown coefficient functions in the model, and large sample properties of the proposed estimators are also obtained. Additionally, an extensive simulation studies are conducted to assess the finite sample performance of the proposed method and a data set from the Academy Awards is analyzed.

Two-Phase, Generalized Case-Control Designs for the Study of Quantitative Longitudinal Outcomes.

We propose a general class of 2-phase epidemiologic study designs for quantitative, longitudinal data that are useful when phase 1 longitudinal outcome and covariate data are available but data on the exposure (e.g., a biomarker) can only be collected on a subset of subjects during phase 2. To conduct a study using a design in the class, one first summarizes the longitudinal outcomes by fitting a simple linear regression of the response on a time-varying covariate for each subject. Sampling strata are defined by splitting the estimated regression intercept or slope distributions into distinct (low, medium, and high) regions. Stratified sampling is then conducted from strata defined by the intercepts, by the slopes, or from a mixture. In general, samples selected with extreme intercept values will yield low variances for associations of time-fixed exposures with the outcome and samples enriched with extreme slope values will yield low variances for associations of time-varying exposures with the outcome (including interactions with time-varying exposures). We describe ascertainment-corrected maximum likelihood and multiple-imputation estimation procedures that permit valid and efficient inferences. We embed all methodological developments within the framework of conducting a substudy that seeks to examine genetic associations with lung function among continuous smokers in the Lung Health Study (United States and Canada, 1986-1994).

The power-law distribution of cumulative coal production

The coal industry is dominated by the largest mines, with 1% of coal mines in the U.S. being responsible for 65% of the coal cumulatively produced. We show that this ”heavy tail” can be well approximated by a power law, where mine-level cumulative production is inversely proportionate to distributional rank. Maximum likelihood and regression-based procedures estimate the counter-cumulative power-law parameter to be less than one, indicating there is no well-defined mean or variance for cumulative production. Goodness of fit tests indicate the power-law is a better fit to the data than other competing distributions, including the lognormal.

Bayesian model selection for the study of Hardy-Weinberg proportions and homogeneity of gender allele frequencies.

Standard statistical tests for Hardy-Weinberg equilibrium assume the equality of allele frequencies in the sexes, whereas tests for the equality of allele frequencies in the sexes assume Hardy-Weinberg equilibrium. This produces a circularity in the testing of genetic variants, which has recently been resolved with new frequentist likelihood and exact procedures. In this paper, we tackle the same problem by posing it as a Bayesian model comparison problem. We formulate an exhaustive set of ten alternative scenarios for biallelic genetic variants. Using Dirichlet and Beta priors for genotype and allele frequencies, we derive marginal likelihoods for all scenarios, and select the most likely scenario using the posterior probabilities that each of these scenarios is the one in place. Different from the usual frequentist testing approach, the Bayesian approach allows one to compare any number of models, and not just two at a time, and the models compared do not have to be nested. We illustrate our Bayesian approach with genetic data from the 1,000 genomes project and through a simulation study.

Odd Hyperbolic Cosine Exponential-Exponential (OHC-EE) Distribution

In the present paper, we introduce a new lifetime distribution based on the general odd hyperbolic cosine-FG model. Some important properties of proposed model including survival function, quantile function, hazard function, order statistic are obtained. In addition estimating unknown parameters of this model will be examined from the perspective of classic and Bayesian statistics. Moreover, an example of real data set is studied; point and interval estimations of all parameters are obtained by maximum likelihood, bootstrap (parametric and non-parametric) and Bayesian procedures. Finally, the superiority of proposed model in terms of parent exponential distribution over other fundamental statistical distributions is shown via the example of real observations.

Genetic parameters for reproductive, longevity and lifetime production traits in Bísaro pigs

The Bísaro pig has gained popularity in recent years reflecting the success of the conservation program. Nevertheless, no data is available for animal genetic evaluation in this breed. Therefore, this study aimed to estimate genetic parameters and trends for reproduction related traits in Bísaro pigs. Through a restricted maximum likelihood procedure applied to mixed linear models, 27,844 farrowing records, from 1995 to 2017, were used to analyse total number of pigs born per litter (NBT), number of pigs born alive (NBA), number of stillborn (NSB), number of pigs weaned per litter (NBW), age at first farrowing (AFF), farrowing interval (FIT), length of productive life (LPL), lifetime number of litters (LNL), lifetime pig production (LTP) and lifetime efficiency (LTP365). The heritability estimates for litter size traits were low and ranged from 0.007 ± 0.004 to 0.015 ± 0.006. Differently, the heritabilities for traits related to longevity and lifetime production traits were higher (0.078 ± 0.026 to 0.121 ± 0.030). AFF registered the highest heritability value (0.345 ± 0.028). NSB and FIT presented high values of additive genetic coefficient of variation (0.177 and 0.271) in contrast with low heritability estimates (0.007 ± 0.004 and 0.002 ± 0.005). Very tight genetic correlations were found between NBT and NBA (0.968), NBW and NBT (0.974), and NBW and NBA (0.945). Weak genetic correlations were found between both NBT and NSB (0.352) and between NBA and NSB (0.107). Longevity and lifetime production traits presented high positive genetic correlations (0.811–0.969) and moderate to high phenotypic correlations (0.266–0.946). No major genetic changes were registered over time for most of the analysed traits, except for AFF and LPL, having registered an overall decrease of mean estimated breeding values (21.3 and 17.5) and negative genetic trends of −0.6 and −0.4 (P < 0.001), respectively.

Internalized stigma and its impact on schizophrenia quality of life

ABSTRACTThe internalization of schizophrenia-related stigma highly affects people living with this disease, hindering their recovery process. The aim of this work is to study the effect of alienation and stereotype endorsement on the specific dimensions of schizophrenia-specific quality of life (SQoL), through the damage on self-esteem and self-efficacy. The model was tested in 167 patients with schizophrenia by path analysis using maximum likelihood procedure. The data fit the proposed model, but not all hypothesized predictions were met. A negative effect of alienation on self-esteem and self-efficacy was shown, while the negative effect of stereotype endorsement on these self-concept-related variables was non-significant. Self-esteem significantly contributed to the variance of the three dimensions of SQoL, while self-efficacy seems not to be involved in the model. Our results suggest that both alienation and stereotype endorsement have a direct effect on psychosocial SQoL and physical SQoL, while motivation and energy-related SQoL is affected only indirectly by alienation (through the decrease of self-esteem). Finally, we found that self-esteem significantly mediates the negative effect of alienation on the three dimensions of SQoL. We propose that the strengthening of self-esteem would mitigate the damage on SQoL, protecting specially the motivation and energy dimension, which may favor recovery.

Negative Parenting Moderates the Prospective Association of ADHD Symptoms and Youth Social Problems.

Although ADHD and negative parenting are established predictors of youth outcomes, their independent and interactive effects on youth social functioning remain unclear. We tested childhood ADHD symptoms and negative parenting as independent and interactive predictors of prospective change in social problems across a four-year follow-up. At baseline, families of 221 (33% female) children with (n = 94) and without ADHD were rigorously assessed including observed positive and negative parenting behavior, youth ADHD symptoms, as well as multi-informant ratings of youth social problems at multiple occasions. Based on multiple regression with robust standard errors and full-information maximum likelihood procedures to address missing data, ADHD symptoms positively predicted social problems, even with control of observed parenting behavior, child age and sex, oppositional defiant disorder symptoms, and baseline social problems. Additionally, a child ADHD symptoms x negative parenting interaction uniquely predicted separate parent- and teacher-rated social problems where ADHD symptoms positively predicted social problems exclusively in the context of high (+1SD) and very high (+2 SD) negative parenting, respectively. When ADHD was separated into distinct dimensions (i.e., inattention, hyperactivity), an interaction between inattention symptoms and negative parenting approached significance such that inattention symptoms positively predicted parent-rated social problems in the context of high negative parenting. We discuss the interaction between parenting and ADHD symptoms in predictions of youth social problems and implications for intervention.

A weighted normalized likelihood procedure for empirical land change modeling

A critical foundation for empirical land change modeling is the mapping of transition potentials—quantitative evaluations of the readiness or suitability of land to go through a transition. This paper presents a procedure based on empirically determined normalized likelihoods of transition. It shows that these normalized likelihoods equate to posterior probabilities if case–control sampling is carried out among historical instances of change and persistence. The posterior probabilities can then be aggregated at the pixel level across multiple covariates using linear opinion pooling where the pixel-specific weight for each covariate is determined locally by its ability to distinguish between the alternatives of change or persistence. Thus, covariates with spatially varying diagnostic ability can be productively incorporated. The resulting algorithm is shown to have a skill comparable to that of a multi-layer perceptron approach with the advantage of high efficiency and amenability to distributed processing in a cloud environment.

Estimation of co-variance components, genetic parameters and genetic trends of VIAscan carcass traits of New Zealand Romney lambs

The aim of this study was to estimate co-variance components, genetic parameters and direct genetic trends of carcass traits of New Zealand Romney lambs assessed by VIAscan technology. Data for VIAscan carcass traits of 3389 lambs that included hot carcass weight (VHW); subcutaneous fat depth near the 12th rib (VGR) and meat yield (expressed as a percentage of VHW in the shoulder (SHY%), loin (LNY%), hind leg (LGY%) and whole carcass (TY%) were analyzed by employing different animal models for univariate and multivariate analysis using restricted maximum likelihood (REML) procedures with ignoring or including maternal effects. Year of birth, parity of ewe and type of birth were included in the model as fixed factors, as well as the linear covariate of age of lamb at drafting was fitted into the model. Based on the most appropriate model, estimates of direct heritability (h2a) ranged from moderate to high for VHW (0.29 ± 0.02), VGR (0.56 ± 0.09), SHY% (0.34 ± 0.07), LNY% (0.37 ± 0.06), LGY% (0.38 ± 0.08) and TY% (0.31 ± 0.02), however, the estimates of maternal heritability (h2m) ranged from low to moderate for VHW (0.10 ± 0.03), VGR (0.38 ± 0.08), SHY% (0.18 ± 0.02), LNY% (0.15 ± 0.04), LGY% (0.19 ± 0.02) and TY% (0.17 ± 0.02). VHW was positively genetically correlated with all VIAscan carcass traits (0.151 to 0.563) except LGY% (−0.145). VGR was positively genetically correlated with both VHW (0.563) and LNY% (0.003), and negatively genetically correlated with the rest of carcass traits (−0.008 to −0.472). The genetic correlation of TY% was negative with VGR (−0.229) and positive with the other carcass traits (0.151 to 0.878). SHY%, LNY% and LGY% were moderately positively genetically correlated with each other with a range (0.538–0.601). Estimating the direct genetic trends of VIAscan carcass traits during the period of study indicated that a high genetic progress was achieved in the desired direction for VHW, LNY%, LGY% and TY%, and in the undesired direction for VGR. The moderate to high heritability estimates for VIAscan carcass traits indicated that these traits could respond to direct selection. Such information is important in designing breeding strategies for improving carcass traits in New Zealand Romney lambs.

Model robust inference with two-stage maximum likelihood estimation for copulas

This article is concerned with inference in parametric copula setups, where both the marginals and the copula have parametric forms. For such models, two-stage maximum likelihood estimation, often referred to as inference function for margins, is used as an attractive alternative to the full maximum likelihood estimation strategy. Previous studies of the two-stage maximum likelihood estimator have largely been based on the assumption that the chosen parametric model captures the true model that generated data. We study the impact of dropping this true model assumption, both theoretically and numerically. We first show that the two-stage maximum likelihood estimator is consistent for a well-defined least false parameter value, different from the analogous least false parameter associated with the full maximum likelihood procedure. Then we demonstrate limiting normality of the full vector of estimators, with concise matrix notation for the variance matrices involved. Along with consistent estimators for these, we have built a model-robust machinery for inference in parametric copula models. The special case where the parametric model is assumed to hold corresponds to situations studied earlier in the literature, with simpler formulas for variance matrices. As a numerical illustration, we perform a set of simulations. We also analyze five-dimensional Norwegian precipitation data. We find that the variance of the copula parameter estimate can both increase and decrease, by dropping the true model assumption. In addition, we observe that the two-stage maximum likelihood estimator is still highly efficient when the true model assumption is dropped and thus the model robust asymptotic variance formulas are used. Additionally, we discover that using highly misspecified models can lead to situations where the asymptotic variance of the two-stage maximum likelihood estimator is lower than that of full maximum likelihood estimator. Our results are also used to analyze the mean squared error properties for both the full and the two-stage maximum likelihood estimators of any focus parameter.

Clothing brand purchase intention through SNS

PurposeThe purpose of this paper is to investigate the intention to purchase products through clothing brands’ social network sites (SNS) based on the theory of planned behavior and uses and gratifications theory (U&G), and the moderating effects of self-image congruity (SIC).Design/methodology/approachData were collected from 1,003 followers of their favorite clothing brands’ SNS. Data analysis was performed using structural equation modeling (SEM) and multi-group SEM analysis. The models were estimated from the matrices of variances and covariances by the maximum likelihood procedure using EQS 6.1.FindingsThe results highlight the positive impact of U&G on attitude, SNS intentions and SNS use, and U&G, SNS intentions and SNS use were seen to be the main antecedents predicting purchase intentions. Furthermore, SIC was found to have moderating effects between SNS attitude and SNS intentions and between SNS intentions and SNS use.Practical implicationsThis research can help clothing brands understand the need to generate brand beliefs, and to develop contents or events to help accomplish the transition from use to purchase.Originality/valueThis research contributes to the literature by providing a better understanding of intention to use and purchase intention through clothing brands’ SNS pages.

Semiparametric inference for a two-stage outcome-dependent sampling design with interval-censored failure time data.

We propose a two-stage outcome-dependent sampling design and inference procedure for studies that concern interval-censored failure time outcomes. This design enhances the study efficiency by allowing the selection probabilities of the second-stage sample, for which the expensive exposure variable is ascertained, to depend on the first-stage observed interval-censored failure time outcomes. In particular, the second-stage sample is enriched by selectively including subjects who are known or observed to experience the failure at an early or late time. We develop a sieve semiparametric maximum pseudo likelihood procedure that makes use of all available data from the proposed two-stage design. The resulting regression parameter estimator is shown to be consistent and asymptotically normal, and a consistent estimator for its asymptotic variance is derived. Simulation results demonstrate that the proposed design and inference procedure performs well in practical situations and is more efficient than the existing designs and methods. An application to a phase 3 HIV vaccine trial is provided.

Complete plastome sequences of 14 African yam species (Dioscorea spp.)

Complete plastomes of 14 African yam species were reconstructed from whole genome sequencing. These plastomes sizes varied from 151,908 to 155,155 bp. We predicted 130 genes, including 18 duplicated genes located in the two inverted regions. Phylogenetic analysis obtained using the maximum likelihood procedure revealed that each species was distinct with high support. This resource will be useful for phylogenetic and diversity analysis of Dioscorea species.

Agronomic evaluation of a Hancornia speciosa Gomes germplasm collection from the Brazilian Cerrado

Hancornia speciosa Gomes is a fruit tree native to Brazil with potential for use in cropping systems. The objective of this study was to characterize accessions of a germplasm collection by evaluating agronomic traits. The collection was planted in 2005, in an experiment with 57 progenies and four replications, totaling 192 individual accessions from 29 natural subpopulations of four botanical varieties, from the Brazilian Cerrado. The components of variance were estimated using the restricted maximum likelihood (REML) procedure. Heritability coefficients show the potential of the collection as base population for selection programs. The trait number of fruits per plant presented the highest estimated selection gain. The accessions of H. speciosa var. cuyabensis are the ones with the greatest potential for selection based on plant size and productivity. Progenies with greater development in the juvenile phase are potentially more productive in the adult phase, allowing the early selection of superior genotypes.