Genetic parameters by Bayesian inference for dual purpose Jaffarabadi buffaloes

Knowledge of genetic parameters is essential for improved reproductive management and increased yield. Quantitative analysis of genetic parameters is lacking for many breeds of buffaloes. This article provides the first estimate of genetic parameters for dual purpose (meat and milk) Brazilian Jaffarabadi buffaloes, using Bayesian inference. Data on milk yield (MY), lactation length (LL), weight at 205 days (W205) and 365 (W365) days of age, and average daily gain (ADG) from 205 to 365 days of age were collected in two herds. Bivariate analyses (using the program MTGSAM) were performed with the Gibbs sampler to obtain estimates of variance and covariance. Average lactation milk yield and lactation length were 1 620.2 +/- 450.9 kg and 257.6 +/- 46.8 days, respectively, and the mean values for weight traits (kg) were 181.6 +/- 63.3 (W205), 298.04 +/- 116.1 (W365), and 0.73 +/- 0.35 (ADG). Heritability estimates (modes) were 0.16 for MY, 0.10 for LL, 0.43 for W205, 0.48 for W365 and 0.32 for ADG. There was a high genetic correlation (0.96) between milk yield and lactation length and very high genetic correlations (0.99) between the three growth traits. Our data suggest that both milk production and growth traits have clear potential for yield improvement through direct selection in this dual purpose breed. The selection for weight at an early age would be successful and selection for MY can be performed in the first lactation.

Genetic Parameters for Reproductive Traits of Crossbred Buffaloes from Brazil, Estimated by Bayesian Inference

The objective of the study was to estimate heritability for calving interval (CI) and age at first calving (AFC) and also calculate repeatability for CI in buffaloes using Bayesian inference. The Brazilian Buffaloes Genetic Improvement Program provided the database. Data consists on information from 628 females and four different herds, born between 1980 and 2003. In order to estimate the variance, univariate analyses were performed employing Gibbs sampler procedure included in the MTGSAM software. The model for CI included the random effects direct additive and permanent environment factors, and the fixed effects of contemporary groups and calving orders. The model for AFC included the direct additive random effect and contemporary groups as a fixed effect. The convergence diagnosis was obtained using Geweke that was implemented through the Bayesian Output Analysis package in R software. The estimated averages were 433.2 days and 36.7months for CI and AFC, respectively. The means, medians and modes for the calculated heritability coefficients were similar. The heritability coefficients were 0.10 and 0.42 for CI and AFC respectively, with a posteriori marginal density that follows a normal distribution for both traits. The repeatability for CI was 0.13. The low heritability estimated for CI indicates that the variation in this trait is, to a large extent, influenced by environmental factors such as herd management policies. The age at first calving has clear potential for yield improvement through direct selection in these animals.

Genetic Correlation for Pre-Weaning and Post-Weaning Traits of Mediterranean Buffaloes from Brazil, Estimated By Bayesian Inference

The aim of this study was to estimate genetic, environmental and phenotypic correlation between birth weight (BW) and weight at 205 days age (W205), BW and weight at 365 days age (W365) and W205-W365, using Bayesian inference. The Brazilian Program for Genetic Improvement of Buffaloes provided the data that included 3,883 observations from Mediterranean breed buffaloes. With the purpose to estimate variance and covariance, bivariate analyses were performed using Gibbs sampler that is included in the MTGSAM software. The model for BW, W205 and W365 included additive direct and maternal genetic random effects, maternal environmental random effect and contemporary group as fixed effect. The convergence diagnosis was achieved using Geweke, a method that uses an algorithm implemented in R software through the package Bayesian Output Analysis. The calculated direct genetic correlations were 0.34 (BW-W205), 0.25 (BW-W365) and 0.74 (W205-W365). The environmental correlations were 0.12, 0.11 and 0.72 between BW-W205, BW-W365 and W205-W365, respectively. The phenotypic correlations were low for BW-W205 (0.01) and BW-W365 (0.04), differently than the obtained for W205-W365 with a value of 0.67. The results indicate that BW trait have low genetic, environmental and phenotypic association with the two others traits. The genetic correlation between W205 and W365 was high and suggests that the selection for weight at around 205 days could be beneficial to accelerate the genetic gain.

Genetic Parameters for Growth Traits of Mediterranean Buffaloes from Brazil, Estimated By Bayesian Inference

Quantitative analysis of growth genetic parameters is not available for many breeds of buffaloes making selection and breeding decisions an empirical process that lacks robustness. The objective of this study was to estimate heritability for birth weight (BW), weight at 205 days (W205) and 365 days (W365) of age using Bayesian inference. The Brazilian Program for Genetic Improvement of Buffaloes provided the data. For the traits BW, W205 and W365 of Brazilian Mediterranean buffaloes 5169, 3792 and 3883 observations have been employed for the analysis, respectively. In order to obtain the estimates of variance, univariate analyses were conducted using the Gibbs sampler included in the MTGSAM software. The model for BW, W205 and W365 included additive direct and maternal genetic random effects, random maternal permanent environmental effect and contemporary group that was treated as a fixed effect. The convergence diagnosis was performed employing Geweke, a method that uses an algorithm from the Bayesian Output Analysis package that was implemented using R software environment. The average values for weight traits were 37.6 +/- 4.7 kg for BW, 192.7 +/- 40.3 kg for W205 and 298.6 +/- 67.4 kg for W365. The heritability posterior distributions for direct and maternal effects were symmetric and close to those expected in a normal distribution. Direct heritability estimates obtained using the modes were 0.30 (BW), 0.52 (W205) and 0.54 (W365). The maternal heritability coefficient estimates were 0.31, 0.19 and 0.21 for BW, W205 and W365, respectively. Our data suggests that all growth traits and mainly W205 and W365, have clear potential for yield improvement through direct genetic selection.

Genetic Parameters for Milk Yield and Lactation Length of Crossbred Buffaloes from Brazil by Bayesian Inference

The objective of the study was to estimate heritability and repeatability for milk yield (MY) and lactation length (LL) in buffaloes using Bayesian inference. The Brazilian genetic improvement program of buffalo provided the data that included 628 females, from four herds, born between 1980 and 2003. In order to obtain the estimates of variance, univariate analyses were performed with the Gibbs sampler, using the MTGSAM software. The model for MY and LL included direct genetic additive and permanent environment as random effects, and contemporary groups, milking frequency and calving number as fixed effects. The convergence diagnosis was performed with the Geweke method using an algorithm implemented in R software through the package Bayesian Output Analysis. Average for milk yield and lactation length was 1,546.1 +/- 483.8 kg and 252.3 +/- 42.5 days, respectively. The heritability coefficients were 0.31 (mode), 0.35 (mean) and 0.34 (median) for MY and 0.11 (mode), 0.10 (mean) and 0.10 (median) for LL. The repeatability coefficient (mode) were 0.50 and 0.15 for MY and LL, respectively. Milk yield is the only trait with clear potential for genetic improvement by direct genetic selection. The repeatability for MY indicates that selection based on the first lactation could contribute for an improvement in this trait.

Bayesian inference for two-parameter gamma distribution assuming different noninformative priors

In this paper distinct prior distributions are derived in a Bayesian inference of the two-parameters Gamma distribution. Noniformative priors, such as Jeffreys, reference, MDIP, Tibshirani and an innovative prior based on the copula approach are investigated. We show that the maximal data information prior provides in an improper posterior density and that the different choices of the parameter of interest lead to different reference priors in this case. Based on the simulated data sets, the Bayesian estimates and credible intervals for the unknown parameters are computed and the performance of the prior distributions are evaluated. The Bayesian analysis is conducted using the Markov Chain Monte Carlo (MCMC) methods to generate samples from the posterior distributions under the above priors.

The zero-inflated Conway-Maxwell-Poisson distribution: Bayesian inference, regression modeling and influence diagnostic

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Essays on Bayesian Inference for Social Networks

This thesis presents Bayesian solutions to inference problems for three types of social network data structures: a single observation of a social network, repeated observations on the same social network, and repeated observations on a social network developing through time. A social network is conceived as being a structure consisting of actors and their social interaction with each other. A common conceptualisation of social networks is to let the actors be represented by nodes in a graph with edges between pairs of nodes that are relationally tied to each other according to some definition. Statistical analysis of social networks is to a large extent concerned with modelling of these relational ties, which lends itself to empirical evaluation. The first paper deals with a family of statistical models for social networks called exponential random graphs that takes various structural features of the network into account. In general, the likelihood functions of exponential random graphs are only known up to a constant of proportionality. A procedure for performing Bayesian inference using Markov chain Monte Carlo (MCMC) methods is presented. The algorithm consists of two basic steps, one in which an ordinary Metropolis-Hastings up-dating step is used, and another in which an importance sampling scheme is used to calculate the acceptance probability of the Metropolis-Hastings step. In paper number two a method for modelling reports given by actors (or other informants) on their social interaction with others is investigated in a Bayesian framework. The model contains two basic ingredients: the unknown network structure and functions that link this unknown network structure to the reports given by the actors. These functions take the form of probit link functions. An intrinsic problem is that the model is not identified, meaning that there are combinations of values on the unknown structure and the parameters in the probit link functions that are observationally equivalent. Instead of using restrictions for achieving identification, it is proposed that the different observationally equivalent combinations of parameters and unknown structure be investigated a posteriori. Estimation of parameters is carried out using Gibbs sampling with a switching devise that enables transitions between posterior modal regions. The main goal of the procedures is to provide tools for comparisons of different model specifications. Papers 3 and 4, propose Bayesian methods for longitudinal social networks. The premise of the models investigated is that overall change in social networks occurs as a consequence of sequences of incremental changes. Models for the evolution of social networks using continuos-time Markov chains are meant to capture these dynamics. Paper 3 presents an MCMC algorithm for exploring the posteriors of parameters for such Markov chains. More specifically, the unobserved evolution of the network in-between observations is explicitly modelled thereby avoiding the need to deal with explicit formulas for the transition probabilities. This enables likelihood based parameter inference in a wider class of network evolution models than has been available before. Paper 4 builds on the proposed inference procedure of Paper 3 and demonstrates how to perform model selection for a class of network evolution models.

Bayesian change-point analysis in linear regression model with scale mixtures of normal distributions

In this thesis, we consider Bayesian inference on the detection of variance change-point models with scale mixtures of normal (for short SMN) distributions. This class of distributions is symmetric and thick-tailed and includes as special cases: Gaussian, Student-t, contaminated normal, and slash distributions. The proposed models provide greater flexibility to analyze a lot of practical data, which often show heavy-tail and may not satisfy the normal assumption. As to the Bayesian analysis, we specify some prior distributions for the unknown parameters in the variance change-point models with the SMN distributions. Due to the complexity of the joint posterior distribution, we propose an efficient Gibbs-type with Metropolis- Hastings sampling algorithm for posterior Bayesian inference. Thereafter, following the idea of [1], we consider the problems of the single and multiple change-point detections. The performance of the proposed procedures is illustrated and analyzed by simulation studies. A real application to the closing price data of U.S. stock market has been analyzed for illustrative purposes.

Data assimilation for precipitation nowcasting using Bayesian inference

This work introduces a new variational Bayes data assimilation method for the stochastic estimation of precipitation dynamics using radar observations for short term probabilistic forecasting (nowcasting). A previously developed spatial rainfall model based on the decomposition of the observed precipitation field using a basis function expansion captures the precipitation intensity from radar images as a set of ‘rain cells’. The prior distributions for the basis function parameters are carefully chosen to have a conjugate structure for the precipitation field model to allow a novel variational Bayes method to be applied to estimate the posterior distributions in closed form, based on solving an optimisation problem, in a spirit similar to 3D VAR analysis, but seeking approximations to the posterior distribution rather than simply the most probable state. A hierarchical Kalman filter is used to estimate the advection field based on the assimilated precipitation fields at two times. The model is applied to tracking precipitation dynamics in a realistic setting, using UK Met Office radar data from both a summer convective event and a winter frontal event. The performance of the model is assessed both traditionally and using probabilistic measures of fit based on ROC curves. The model is shown to provide very good assimilation characteristics, and promising forecast skill. Improvements to the forecasting scheme are discussed

Bayesian Inference in Large-scale Problems

Many modern applications fall into the category of "large-scale" statistical problems, in which both the number of observations n and the number of features or parameters p may be large. Many existing methods focus on point estimation, despite the continued relevance of uncertainty quantification in the sciences, where the number of parameters to estimate often exceeds the sample size, despite huge increases in the value of n typically seen in many fields. Thus, the tendency in some areas of industry to dispense with traditional statistical analysis on the basis that "n=all" is of little relevance outside of certain narrow applications. The main result of the Big Data revolution in most fields has instead been to make computation much harder without reducing the importance of uncertainty quantification. Bayesian methods excel at uncertainty quantification, but often scale poorly relative to alternatives. This conflict between the statistical advantages of Bayesian procedures and their substantial computational disadvantages is perhaps the greatest challenge facing modern Bayesian statistics, and is the primary motivation for the work presented here.

Two general strategies for scaling Bayesian inference are considered. The first is the development of methods that lend themselves to faster computation, and the second is design and characterization of computational algorithms that scale better in n or p. In the first instance, the focus is on joint inference outside of the standard problem of multivariate continuous data that has been a major focus of previous theoretical work in this area. In the second area, we pursue strategies for improving the speed of Markov chain Monte Carlo algorithms, and characterizing their performance in large-scale settings. Throughout, the focus is on rigorous theoretical evaluation combined with empirical demonstrations of performance and concordance with the theory.

One topic we consider is modeling the joint distribution of multivariate categorical data, often summarized in a contingency table. Contingency table analysis routinely relies on log-linear models, with latent structure analysis providing a common alternative. Latent structure models lead to a reduced rank tensor factorization of the probability mass function for multivariate categorical data, while log-linear models achieve dimensionality reduction through sparsity. Little is known about the relationship between these notions of dimensionality reduction in the two paradigms. In Chapter 2, we derive several results relating the support of a log-linear model to nonnegative ranks of the associated probability tensor. Motivated by these findings, we propose a new collapsed Tucker class of tensor decompositions, which bridge existing PARAFAC and Tucker decompositions, providing a more flexible framework for parsimoniously characterizing multivariate categorical data. Taking a Bayesian approach to inference, we illustrate empirical advantages of the new decompositions.

Latent class models for the joint distribution of multivariate categorical, such as the PARAFAC decomposition, data play an important role in the analysis of population structure. In this context, the number of latent classes is interpreted as the number of genetically distinct subpopulations of an organism, an important factor in the analysis of evolutionary processes and conservation status. Existing methods focus on point estimates of the number of subpopulations, and lack robust uncertainty quantification. Moreover, whether the number of latent classes in these models is even an identified parameter is an open question. In Chapter 3, we show that when the model is properly specified, the correct number of subpopulations can be recovered almost surely. We then propose an alternative method for estimating the number of latent subpopulations that provides good quantification of uncertainty, and provide a simple procedure for verifying that the proposed method is consistent for the number of subpopulations. The performance of the model in estimating the number of subpopulations and other common population structure inference problems is assessed in simulations and a real data application.

In contingency table analysis, sparse data is frequently encountered for even modest numbers of variables, resulting in non-existence of maximum likelihood estimates. A common solution is to obtain regularized estimates of the parameters of a log-linear model. Bayesian methods provide a coherent approach to regularization, but are often computationally intensive. Conjugate priors ease computational demands, but the conjugate Diaconis--Ylvisaker priors for the parameters of log-linear models do not give rise to closed form credible regions, complicating posterior inference. In Chapter 4 we derive the optimal Gaussian approximation to the posterior for log-linear models with Diaconis--Ylvisaker priors, and provide convergence rate and finite-sample bounds for the Kullback-Leibler divergence between the exact posterior and the optimal Gaussian approximation. We demonstrate empirically in simulations and a real data application that the approximation is highly accurate, even in relatively small samples. The proposed approximation provides a computationally scalable and principled approach to regularized estimation and approximate Bayesian inference for log-linear models.

Another challenging and somewhat non-standard joint modeling problem is inference on tail dependence in stochastic processes. In applications where extreme dependence is of interest, data are almost always time-indexed. Existing methods for inference and modeling in this setting often cluster extreme events or choose window sizes with the goal of preserving temporal information. In Chapter 5, we propose an alternative paradigm for inference on tail dependence in stochastic processes with arbitrary temporal dependence structure in the extremes, based on the idea that the information on strength of tail dependence and the temporal structure in this dependence are both encoded in waiting times between exceedances of high thresholds. We construct a class of time-indexed stochastic processes with tail dependence obtained by endowing the support points in de Haan's spectral representation of max-stable processes with velocities and lifetimes. We extend Smith's model to these max-stable velocity processes and obtain the distribution of waiting times between extreme events at multiple locations. Motivated by this result, a new definition of tail dependence is proposed that is a function of the distribution of waiting times between threshold exceedances, and an inferential framework is constructed for estimating the strength of extremal dependence and quantifying uncertainty in this paradigm. The method is applied to climatological, financial, and electrophysiology data.

The remainder of this thesis focuses on posterior computation by Markov chain Monte Carlo. The Markov Chain Monte Carlo method is the dominant paradigm for posterior computation in Bayesian analysis. It has long been common to control computation time by making approximations to the Markov transition kernel. Comparatively little attention has been paid to convergence and estimation error in these approximating Markov Chains. In Chapter 6, we propose a framework for assessing when to use approximations in MCMC algorithms, and how much error in the transition kernel should be tolerated to obtain optimal estimation performance with respect to a specified loss function and computational budget. The results require only ergodicity of the exact kernel and control of the kernel approximation accuracy. The theoretical framework is applied to approximations based on random subsets of data, low-rank approximations of Gaussian processes, and a novel approximating Markov chain for discrete mixture models.

Data augmentation Gibbs samplers are arguably the most popular class of algorithm for approximately sampling from the posterior distribution for the parameters of generalized linear models. The truncated Normal and Polya-Gamma data augmentation samplers are standard examples for probit and logit links, respectively. Motivated by an important problem in quantitative advertising, in Chapter 7 we consider the application of these algorithms to modeling rare events. We show that when the sample size is large but the observed number of successes is small, these data augmentation samplers mix very slowly, with a spectral gap that converges to zero at a rate at least proportional to the reciprocal of the square root of the sample size up to a log factor. In simulation studies, moderate sample sizes result in high autocorrelations and small effective sample sizes. Similar empirical results are observed for related data augmentation samplers for multinomial logit and probit models. When applied to a real quantitative advertising dataset, the data augmentation samplers mix very poorly. Conversely, Hamiltonian Monte Carlo and a type of independence chain Metropolis algorithm show good mixing on the same dataset.

Hierarchical Task Network Planning with Common-Sense Reasoning for Multiple-People Behaviour Analysis

Safety on public transport is a major concern for the relevant authorities. We
address this issue by proposing an automated surveillance platform which combines data from video, infrared and pressure sensors. Data homogenisation and integration is achieved by a distributed architecture based on communication middleware that resolves interconnection issues, thereby enabling data modelling. A common-sense knowledge base models and encodes knowledge about public-transport platforms and the actions and activities of passengers. Trajectory data from passengers is modelled as a time-series of human activities. Common-sense knowledge and rules are then applied to detect inconsistencies or errors in the data interpretation. Lastly, the rationality that characterises human behaviour is also captured here through a bottom-up Hierarchical Task Network planner that, along with common-sense, corrects misinterpretations to explain passenger behaviour. The system is validated using a simulated bus saloon scenario as a case-study. Eighteen video sequences were recorded with up to six passengers. Four metrics were used to evaluate performance. The system, with an accuracy greater than 90% for each of the four metrics, was found to outperform a rule-base system and a system containing planning alone.

Mechanical behaviour analysis of polyester polymer mortars modified with recycled GFRP waste materials

In this study the effect of incorporation of recycled glass-fibre reinforced polymer (GFRP) waste materials, obtained by means of milling processes, on mechanical behaviour of polyester polymer mortars was assessed. For this purpose, different contents of recycled GFRP waste powder and fibres, with distinct size gradings, were incorporated into polyester based mortars as sand aggregates and filler replacements. Flexural and compressive loading capacities were evaluated and found better than unmodified polymer mortars. GFRP modified polyester based mortars also show a less brittle behaviour, with retention of some loading capacity after peak load. Obtained results highlight the high potential of recycled GFRP waste materials as efficient and sustainable reinforcement and admixture for polymer concrete and mortars composites, constituting an emergent waste management solution.

Bayesian Inference in the Time Varying Cointegration Model

There are both theoretical and empirical reasons for believing that the parameters of macroeconomic models may vary over time. However, work with time-varying parameter models has largely involved Vector autoregressions (VARs), ignoring cointegration. This is despite the fact that cointegration plays an important role in informing macroeconomists on a range of issues. In this paper we develop time varying parameter models which permit cointegration. Time-varying parameter VARs (TVP-VARs) typically use state space representations to model the evolution of parameters. In this paper, we show that it is not sensible to use straightforward extensions of TVP-VARs when allowing for cointegration. Instead we develop a specification which allows for the cointegrating space to evolve over time in a manner comparable to the random walk variation used with TVP-VARs. The properties of our approach are investigated before developing a method of posterior simulation. We use our methods in an empirical investigation involving a permanent/transitory variance decomposition for inflation.

Cancer mortality inequalities in urban areas: a Bayesian small area analysis in Spanish cities

Background Intra-urban inequalities in mortality have been infrequently analysed in European contexts. The aim of the present study was to analyse patterns of cancer mortality and their relationship with socioeconomic deprivation in small areas in 11 Spanish cities. Methods It is a cross-sectional ecological design using mortality data (years 1996-2003). Units of analysis were the census tracts. A deprivation index was calculated for each census tract. In order to control the variability in estimating the risk of dying we used Bayesian models. We present the RR of the census tract with the highest deprivation vs. the census tract with the lowest deprivation. Results In the case of men, socioeconomic inequalities are observed in total cancer mortality in all cities, except in Castellon, Cordoba and Vigo, while Barcelona (RR = 1.53 95%CI 1.42-1.67), Madrid (RR = 1.57 95%CI 1.49-1.65) and Seville (RR = 1.53 95%CI 1.36-1.74) present the greatest inequalities. In general Barcelona and Madrid, present inequalities for most types of cancer. Among women for total cancer mortality, inequalities have only been found in Barcelona and Zaragoza. The excess number of cancer deaths due to socioeconomic deprivation was 16,413 for men and 1,142 for women. Conclusion This study has analysed inequalities in cancer mortality in small areas of cities in Spain, not only relating this mortality with socioeconomic deprivation, but also calculating the excess mortality which may be attributed to such deprivation. This knowledge is particularly useful to determine which geographical areas in each city need intersectorial policies in order to promote a healthy environment.