Search

Robust inferential methods based on divergences measures have shown an appealing trade-off between efficiency and robustness in many different statistical models. In this paper, minimum density power divergence estimators (MDPDEs) for the scale and shape parameters of the log-logistic distribution are considered. The log-logistic is a versatile distribution modeling lifetime data which is commonly adopted in survival analysis and reliability engineering studies when the hazard rate is initially increasing but then it decreases after some point. Further, it is shown that the classical estimators based on maximum likelihood (MLE) are included as a particular case of the MDPDE family. Moreover, the corresponding influence function of the MDPDE is obtained, and its boundlessness is proved, thus leading to robust estimators. A simulation study is carried out to illustrate the slight loss in efficiency of MDPDE with respect to MLE and, at besides, the considerable gain in robustness.

In this paper we apply divergence measures to empirical likelihood applied to logistic regression models. We define a family of empirical test statistics based on divergence measures, called empirical phi-divergence test statistics, extending the empirical likelihood ratio test. We study the asymptotic distribution of these empirical test statistics, showing that it is the same for all the test statistics in this family, and the same as the classical empirical likelihood ratio test. Next, we study the power function for the members in this family, showing that the empirical phi-divergence tests introduced in the paper are consistent in the Fraser sense. In order to compare the differences in behavior among the empirical phi-divergence test statistics in this new family, considered for the first time in this paper, we carry out a simulation study.

In this paper, we develop robust estimators and tests for one-shot device testing under proportional hazards assumption based on divergence measures. Through a detailed Monte Carlo simulation study and a numerical example, the developed inferential procedures are shown to be more robust than the classical procedures, based on maximum likelihood estimators.

Most work on one-shot devices assume that there is only one possible cause of device failure. However, in practice, it is often the case that the products under study can experience any one of various possible causes of failure. Robust estimators and Wald-type tests are developed here for the case of one-shot devices under competing risks. An extensive simulation study illustrates the robustness of these divergence-based estimators and test procedures based on them. A data-driven procedure is proposed for choosing the optimal estimator for any given data set which is then applied to an example in the context of survival analysis.

We present a dynamic and thermodynamic study of the orientational glass former Freon113 (CCl2F-CClF2) in order to analyze its kinetic and thermodynamic fragilities. Freon113 displays internal molecular degrees of freedom which promote a complex energy landscape. Experimental specific heat and its microscopic origin, the vibrational density of states from inelastic neutron scattering, together with the orientational dynamics obtained by means of dielectric spectroscopy have revealed the highest fragility value, both thermodynamic and kinetic, found for this orientational glass former. The excess in both Debye-reduced specific heat and density of states (boson peak) evidences the existence of glassy low-energy excitations. We demonstrate that early proposed correlations between the boson peak and the Debye specific heat value are elusive as revealed by the clear counterexample of the studied case., Comment: 13 pages, 5 figures

This paper develops a new family of estimators, the minimum density power divergence estimators (MDPDEs), for the parameters of the one-shot device model as well as a new family of test statistics, Z-type test statistics based on MDPDEs, for testing the corresponding model parameters. The family of MDPDEs contains as a particular case the maximum likelihood estimator (MLE) considered in Balakrishnan and Ling (2012). Through a simulation study, it is shown that some MDPDEs have a better behavior than the MLE in relation to robustness. At the same time, it can be seen that some Z-type tests based on MDPDEs have a better behavior than the classical Z-test statistic also in terms of robustness.

This paper derives a new family of estimators, namely the minimum density power divergence estimators, as a robust generalization of the maximum likelihood estimator for the polytomous logistic regression model. Based on these estimators, a family of Wald-type test statistics for linear hypotheses is introduced. Robustness properties of both the proposed estimators and the test statistics are theoretically studied through the classical influence function analysis. Appropriate real life examples are presented to justify the requirement of suitable robust statistical procedures in place of the likelihood based inference for the polytomous logistic regression model. The validity of the theoretical results established in the paper are further confirmed empirically through suitable simulation studies. Finally, an approach for the data-driven selection of the robustness tuning parameter is proposed with empirical justifications., Comment: Some typos were corrected from the published version

Fitting a data set with a parametrized model can be seen geometrically as finding the global minimum of the chi^2 hypersurface, depending on a set of parameters {P_i}. This is usually done using the Levenberg-Marquardt algorithm. The main drawback of this algorithm is that despite of its fast convergence, it can get stuck if the parameters are not initialized close to the final solution. We propose a modification of the Metropolis algorithm introducing a parameter step tuning that optimizes the sampling of parameter space. The ability of the parameter tuning algorithm together with simulated annealing to find the global chi^2 hypersurface minimum, jumping across chi^2{P_i} barriers when necessary, is demonstrated with synthetic functions and with real data.

We present a study of the dynamics and structural changes for trans-1,2-dichloroethylene between high and low density liquids using neutron scattering techniques (diffraction, small angle neutron scattering and time of flight spectroscopy) and molecular dynamics simulations. We show that changes in the short range ordering of molecules goes along with a change of the molecular dynamics: both structure and dynamics of the high density liquid are more cooperative than those of the low density liquid. The microscopic mechanism underlying the cooperative motions in the high density liquid has been found to be related to the backscattering of molecules due to a strong correlation of molecular ordering., Comment: 5 pages and 3 figures. To be published in Physical Review B

The extraction of any physical information from quasielastic neutron scattering spectra is generally done by fitting a model to the data by means of chi-square minimization procedure. However, as pointed out by the pioneering work of D.S. Sivia et al., also another probabilistic approach based on Bayes theorem can be employed. In a nutshell, the main difference between the classical chi-square minimization and the Bayesian approach is the way of expressing the final results: In the first case, the result is a set of values of parameters with a symmetric error and a figure of merit such as chi-square, whereas in the second case the results are presented as probability distribution functions (PDF) of both, parameters and merit figure. In this contribution, we demonstrate how final PDFs are obtained by exploring all possible combinations of parameters that are compatible with the experimental error. Three advantages of this method will be emphasized: First, correlations between parameters are automatically taken into account, which implies, for example, that parameter errors are correctly calculated, correlations show up in a natural way and ill defined parameters are immediately recognized from their PDF. Second, it is possible to calculate the likelihood of a determined physical model, and therefore to select the one among many that fits the data best with a minimal number of parameters, in a correctly defined probabilistic way.

We have performed a detailed dielectric investigation of the relaxational dynamics of glass-forming benzophenone. Our measurements cover a broad frequency range of 0.1 Hz to 120 GHz and temperatures from far below the glass temperature well up into the region of the small-viscosity liquid. With respect to the alpha relaxation this material can be characterized as a typical molecular glass former with rather high fragility. A good agreement of the alpha relaxation behavior with the predictions of the mode coupling theory of the glass transition is stated. In addition, at temperatures below and in the vicinity of Tg we detect a well-pronounced beta relaxation of Johari-Goldstein type, which with increasing temperature develops into an excess wing. We compare our results to literature data from optical Kerr effect and depolarized light scattering experiments, where an excess-wing like feature was observed in the 1 - 100 GHz region. We address the question if the Cole-Cole peak, which was invoked to describe the optical Kerr effect data within the framework of the mode coupling theory, has any relation to the canonical beta relaxation detected by dielectric spectroscopy., Comment: 11 pages, 7 figures; revised version with new Fig. 5 and some smaller changes according to referees' demands

We report a thorough characterization of the glassy dynamics of benzophenone by broadband dielectric spectroscopy. We detect a well pronounced beta relaxation peak developing into an excess wing with increasing temperature. A previous analysis of results from Optical-Kerr-effect measurements on this material within the mode coupling theory revealed a high-frequency Cole-Cole peak. We address the question if this phenomenon also may explain the Johari-Goldstein beta relaxation, a so far unexplained spectral feature inherent to glass-forming matter, mainly observed in dielectric spectra. Our results demonstrate that according to the present status of theory, both spectral features seem not to be directly related., Comment: 4 pages, 3 figures; second version as published; small changes of text according to referee suggestions; title changed according to editor's demands

We present a thorough dielectric investigation of the relaxation dynamics of plastic crystalline Freon112, which exhibits freezing of the orientational degrees of freedom into a glassy crystal below 90 K. Among other plastic crystals, Freon112 stands out by being relatively fragile within Angell's classification scheme and by showing an unusually strong $\beta$-relaxation. Comparing the results to those on Freon112a, having only a single molecular conformation, points to the importance of the presence of two molecular conformations in Freon112 for the explanation of its unusual properties., Comment: 17 pages, 6 figures

We show that rational data of bounded input length are uniformly distributed with respect to condition numbers of numerical analysis. We deal both with condition numbers of Linear Algebra and with condition numbers for systems of multivariate polynomial equations. For instance, we show that for any $w>1$ and for any $n\times n$ rational matrix $M$ of bit length $O(n^4\log n) + \log w$, the condition number $k(M)$ satisfies $k(M) \leq w n^{5/2}$ with probability at least $1-2w^{-1}$. Similar estimates are shown for the condition number $\mu_{norm}$ of M. Shub and S. Smale when applied to systems of multivariate homogeneous polynomial equations of bounded input length. Finally we apply these techniques to show the probability distribution of the precision (number of bits of the denominator) required to write down approximate zeros of affine systems of multivariate polynomial equations of bounded input length., Comment: 54 pages

In this extended abstract we deal with the relations between the numerical/diophantine approximation and the symbolic/algebraic geometry approachs to solving of multivariate diophentine polynomial systems, obtaining several consecuences ranging from diophantine approximation to effective number theory.

This paper is devoted to the complexity analysis of a particular property, called "algebraic robustness" owned by all known symbolic methods of parametric polynomial equation solving (geometric elimination). It is shown that any parametric elimination procedure which owns this property must neccessarily have an exponential sequential time complexity., Comment: LaTeX, 12 pages. To appear in Proc. of WAIT'97, Buenos Aires, Argentina, 1997

We present a new method for solving symbolically zero--dimensional polynomial equation systems in the affine and toric case. The main feature of our method is the use of problem adapted data structures: arithmetic networks and straight--line programs. For sequential time complexity measured by network size we obtain the following result: it is possible to solve any affine or toric zero--dimensional equation system in non--uniform sequential time which is polynomial in the length of the input description and the ``geometric degree" of the equation system. Here, the input is thought to be given by a straight--line program (or alternatively in sparse representation), and the length of the input is measured by number of variables, degree of equations and size of the program (or sparsity of the equations). The geometric degree of the input system has to be adequately defined. It is always bounded by the algebraic--combinatoric "B\'ezout number" of the system which is given by the Hilbert function of a suitable homogeneous ideal. However, in many important cases, the value of the geometric degree of the system is much smaller than its B\'ezout number since this geometric degree does not take into account multiplicities or degrees of extraneous components (which may appear at infinity in the affine case or may be contained in some coordinate hyperplane in the toric case). Our method contains a new application of a classic tool to symbolic computation: we use Newton iteration in order to simplify straight--line programs occurring in elimination procedures. Our new technique allows for practical implementations a meaningful characterization of the intrinsic {\it algebraic complexity} of typic elimination problems and reduces the still unanswered question of their intrinsic {\it bit complexity} to, Comment: Latex. To appear in Journal of Pure and Applied Algebra

We introduce a subexponential algorithm for geometric solving of multivariate polynomial equation systems whose bit complexity depends mainly on intrinsic geometric invariants of the solution set. From this algorithm, we derive a new procedure for the decision of consistency of polynomial equation systems whose bit complexity is subexponential, too. As a byproduct, we analyze the division of a polynomial modulo a reduced complete intersection ideal and from this, we obtain an intrinsic lower bound for the logarithmic height of diophantine approximations to a given solution of a zero--dimensional polynomial equation system. This result represents a multivariate version of Liouville's classical theorem on approximation of algebraic numbers by rationals. A special feature of our procedures is their {\em polynomial} character with respect to the mentioned geometric invariants when instead of bit operations only arithmetic operations are counted at unit cost. Technically our paper relies on the use of straight--line programs as a data structure for the encoding of polynomials, on a new symbolic application of Newton's algorithm to the Implicit Function Theorem and on a special, basis independent trace formula for affine Gorenstein algebras.