Comparison of approaches for parameter estimation on stochastic models: Generic least squares versus specialized approaches

Parameter estimation for models with intrinsic stochasticity poses specific challenges that do not exist for deterministic models. Therefore, specialized numerical methods for parameter estimation in stochastic models have been developed. Here, we study whether dedicated algorithms for stochastic models are indeed superior to the naive approach of applying the readily available least squares algorithm designed for deterministic models.We compare the performance of the recently developed multiple shooting for stochastic systems (MSS) method designed for parameter estimation in stochastic models, a stochastic differential equations based Bayesian approach and a chemical master equation based techniques with the least squares approach for parameter estimation in models of ordinary differential equations (ODE). As test data, 1000 realizations of the stochastic models are simulated. For each realization an estimation is performed with each method, resulting in 1000 estimates for each approach. These are compared with respect to their deviation to the true parameter and, for the genetic toggle switch, also their ability to reproduce the symmetry of the switching behavior. Results are shown for different set of parameter values of a genetic toggle switch leading to symmetric and asymmetric switching behavior as well as an immigration-death and a susceptible-infected-recovered model. This comparison shows that it is important to choose a parameter estimation technique that can treat intrinsic stochasticity and that the specific choice of this algorithm shows only minor performance differences.     

On the Bayesian Mixture Model and Identifiability

This article is concerned with Bayesian mixture models and identifiability issues. There are two sources of unidentifiability: the well-known likelihood invariance under label switching and the perhaps less well-known parameter identifiability problem. When using latent allocation variables determined by the mixture model, these sources of unidentifiability create arbitrary labeling that renders estimation of the model very difficult. We endeavor to tackle these problems by proposing a prior distribution on the allocations, which provides an explicit interpretation for the labeling by removing gaps with high probability. We propose a Markov chain Monte Carlo (MCMC) estimation method and present supporting illustrations.     

Robust state estimation and fault diagnosis for uncertain hybrid systems

Robust state estimation and fault diagnosis are challenging problems in the research into hybrid systems. In this paper a novel robust hybrid observer is proposed for a class of hybrid systems with unknown inputs and faults. Model uncertainties, disturbances and faults are represented as structured unknown inputs. The robust hybrid observer consists of a mode observer for mode identification and a continuous observer for continuous state estimation and mode transition detection. It is shown that the mode can be identified correctly and the continuous state estimation error is exponentially uniformly bounded. Robustness to model uncertainties and disturbances can be guaranteed for the hybrid observer by disturbance decoupling. Furthermore, the detectability and mode identifiability conditions are rigorously analyzed. On the basis of the robust hybrid observer, a robust fault detection and isolation scheme is presented also in the paper. Simulations of a hybrid four-tank system show the proposed approach is effective.     

Rational reconstruction of frailty-based mortality models by a generalisation of Gompertz’ law of mortality

A generalisation of Gompertz’ distribution is proposed, and it is shown that continuous heterogeneous mortality models with Gamma distributed frailty have lifetime random variables distributed as the difference of two such generalised Gompertz random variables. With this result, limitations of existing frailty-based mortality models are identified. The approach taken in this paper allows the frailty distribution to be interpreted as a lifetime reduction distribution and enables application of heterogeneous survival models with a stronger relation to empirically identifiable concepts.     

Identifiability analysis of an epidemiological model in a structured population

We investigate the parameter identifiability problem for a system of nonlinear integro-partial differential equations of transport type, representing the spread of a disease with a long infectious but undetectable period in an individual population. After obtaining the expression of the model input-output relationships, we give sufficient conditions on the initial and boundary conditions of the system that guarantee the parameter identifiability on a finite time horizon. We finally illustrate our findings with numerical simulations.     

Nonparametric Estimation of Distributions in Random Effects Models

We propose using minimum distance to obtain nonparametric estimates of the distributions of components in random effects models. A main setting considered is equivalent to having a large number of small datasets whose locations, and perhaps scales, vary randomly, but which otherwise have a common distribution. Interest focuses on estimating the distribution that is common to all datasets, knowledge of which is crucial in multiple testing problems where a location/scale invariant test is applied to every small dataset. A detailed algorithm for computing minimum distance estimates is proposed, and the usefulness of our methodology is illustrated by a simulation study and an analysis of microarray data. Supplemental materials for the article, including R-code and a dataset, are available online.     

Label Switching in Bayesian Mixture Models: Deterministic Relabeling Strategies

Label switching is a well-known problem in the Bayesian analysis of mixture models. On the one hand, it complicates inference, and on the other hand, it has been perceived as a prerequisite to justify Markov chain Monte Carlo (MCMC) convergence. As a result, nonstandard MCMC algorithms that traverse the symmetric copies of the posterior distribution, and possibly genuine modes, have been proposed. To perform component-specific inference, methods to undo the label switching and to recover the interpretation of the components need to be applied. If latent allocations for the design of the MCMC strategy are included, and the sampler has converged, then labels assigned to each component may change from iteration to iteration. However, observations being allocated together must remain similar, and we use this fundamental fact to derive an easy and efficient solution to the label switching problem. We compare our strategy with other relabeling algorithms on univariate and multivariate data examples and demonstrate improvements over alternative strategies. Supplementary materials for this article are available online.     

On Huber's contaminated model

Huber's contaminated model is a basic model for data with outliers. This paper aims at addressing several fundamental problems about this model. We first study its identifiability properties. Several theorems are presented to determine whether the model is identifiable for various situations. Based on these results, we discuss the problem of estimating the parameters with observations drawn from Huber's contaminated model. A definition of estimation consistency is introduced to handle the general case where the model may be unidentifiable. This consistency is a strong robustness property. After showing that existing estimators cannot be consistent in this sense, we propose a new estimator that possesses the consistency property under mild conditions. Its adaptive version, which can simultaneously possess this consistency property and optimal asymptotic efficiency, is also provided. Numerical examples show that our estimators have better overall performance than existing estimators no matter how many outliers in the data.     

A note on the identifiability of nonparametric and semiparametric mixtures of GLMs

In this article, we first propose a semiparametric mixture of generalized linear models (GLMs) and a nonparametric mixture of GLMs, and then establish identifiability results under mild conditions.