arXiv subjects
Publications and source records attributed to N. Balakrishnan.
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.
Quantum scattering calculations for all but low-dimensional systems at low energies must rely on approximations. All approximations introduce errors. The impact of these errors is often difficult to assess because they depend on the Hamiltonian parameters and the particular observable under study. Here, we illustrate a general, system and approximation-independent, approach to improve the accuracy of quantum dynamics approximations. The method is based on a Bayesian machine learning (BML) algorithm that is trained by a small number of rigorous results and a large number of approximate calculations, resulting in ML models that accurately capture the dependence of the dynamics results on the quantum dynamics parameters. Most importantly, the present work demonstrates that the BML models can generalize quantum results to different dynamical processes. Thus, a ML model trained by a combination of approximate and rigorous results for a certain inelastic transition can make accurate predictions for different transitions without rigorous calculations. This opens the possibility of improving the accuracy of approximate calculations for quantum transitions that are out of reach of rigorous scattering calculations.
Conditional specification of distributions is a developing area with many applications. In the finite discrete case, a variety of compatible conditions can be derived. In this paper, we propose an alternative approach to study the compatibility of two conditional probability distributions under the finite discrete set up. A technique based on rank-based criterion is shown to be particularly convenient for identifying compatible distributions corresponding to complete conditional specification, including the case with zeros. The proposed methods are finally illustrated with several examples.
In this paper, we propose a closed form approximation to the mean and variance of a new generalization of negative binomial (NGNB) distribution arising from the Extended COM-Poisson (ECOMP) distribution developed by Chakraborty and Imoto (2016)(see [4]). The NGNB is a special case of the ECOMP distribution and was named so by these authors. This distribution is more flexible in terms of the dispersion index as compared to its ordinary counterparts. It approaches the COM-Poisson distribution (Shmueli et al. 2005) [11] under suitable limiting conditions. The NGNB can also be obtained from the COM-Negative Hypergeometric distribution (Roy et al. 2019)[10] as a limiting distribution. In this paper, we present closed-form approximations for the mean and variance of the NGNB distribution. These approximations can be viewed as the mean and variance of convolution of independent and identically distributed negative binomial populations. The proposed closed-form approximations of the mean and variance will be helpful in building the link function for the generalized negative binomial regression model based on the NGNB distribution and other extended applications, hence resulting in enhanced applicability of this model.
The relative orientation of colliding molecules plays a key role in determining the rates of chemical processes. Here we examine in detail a prototypical example: rotational quenching of HD in cold collisions with H2. We show that the rotational quenching rate from j=2 -> 0, in the v=1 vibrational level, can be maximized by aligning the HD along the collision axis and can be minimized by aligning the HD at the so called magic angle. This follows from quite general helicity considerations and suggests that quenching rates for other similar systems can also be controlled in this manner.
Birnbaum and Saunders introduced a two-parameter lifetime distribution to model fatigue life of a metal, subject to cyclic stress. Since then, extensive work has been done on this model providing different interpretations, constructions, generalizations, inferential methods, and extensions to bivariate, multivariate and matrix-variate cases. More than two hundred papers and one research monograph have already appeared describing all these aspects and developments. In this paper, we provide a detailed review of all these developments and at the same time indicate several open problems that could be considered for further research.
A common assumption regarding the standard tobit model is the normality of the error distribution. However, asymmetry and bimodality may be present and alternative tobit models must be used. In this paper, we propose a tobit model based on the class of log-symmetric distributions, which includes as special cases heavy and light tailed distributions and bimodal distributions. We implement a likelihood-based approach for parameter estimation and derive a type of residual. We then discuss the problem of performing testing inference in the proposed class by using the likelihood ratio and gradient statistics, which are particularly convenient for tobit models, as they do not require the information matrix. A thorough Monte Carlo study is presented to evaluate the performance of the maximum likelihood estimators and the likelihood ratio and gradient tests. Finally, we illustrate the proposed methodology by using a real-world data set.
We report the first full-dimensional potential energy surface (PES) and quantum mechanical close-coupling calculations for scattering of SiO due to H$_2$. The full-dimensional interaction potential surface was computed using the explicitly correlated coupled-cluster (CCSD(T)-F12b) method and fitted using an invariant polynomial approach. Pure rotational quenching cross sections from initial states $v_1=0$, $j_1$=1-5 of SiO in collision with H$_2$ are calculated for collision energies between 1.0 and 5000 cm$^{-1}$. State-to-state rotational rate coefficients are calculated at temperatures between 5 and 1000 K. The rotational rate coefficients of SiO with para-H$_2$ are compared with previous approximate results which were obtained using SiO-He PESs or scaled from SiO-He rate coefficients. Rovibrational state-to-state and total quenching cross sections and rate coefficients for initially excited SiO($v_1=1, j_1$=0 and 1) in collisions with para-H$_2$($v_2=0,j_2=0$) and ortho-H$_2$($v_2=0,j_2=1$) were also obtained. The application of the current collisional rate coefficients to astrophysics is briefly discussed.
In this paper, we present a fractional decomposition of the probability generating function of the innovation process of the first-order non-negative integer-valued autoregressive [INAR(1)] process to obtain the corresponding probability mass function. We also provide a comprehensive review of integer-valued time series models, based on the concept of thinning operators, with geometric-type marginals. In particular, we develop four fractional approaches to obtain the distribution of innovation processes of the INAR(1) model and show that the distribution of the innovations sequence has geometric-type distribution. These approaches are discussed in detail and illustrated through a few examples. Finally, using the methods presented here, we develop four new first-order non-negative integer-valued autoregressive process for autocorrelated counts with overdispersion with known marginals, and derive some properties of these models.
Lifetimes of complexes formed during ultracold collisions are of current experimental interest as a possible cause of trap loss in ultracold gases of alkali-dimers. Microsecond lifetimes for complexes formed during ultracold elastic collisions of K2 with Rb are reported, from numerically-exact quantum-scattering calculations. The reported lifetimes are compared with those calculated using a simple density-of-states approach, which are shown to be reasonable. Long-lived complexes correspond to narrow scattering resonances which we examine for the statistical signatures of quantum chaos, finding that the positions and widths of the resonances follow the Wigner-Dyson and Porter-Thomas distributions respectively.
A fundamental question in the study of chemical reactions is how reactions proceed at a collision energy close to absolute zero. This question is no longer hypothetical: quantum degenerate gases of atoms and molecules can now be created at temperatures lower than a few tens of nanoKelvin. In this work we consider the benchmark ultracold reaction between, the most-celebrated ultracold molecule, KRb and K. For the first time we map out an accurate ab initio ground state potential energy surface of the KRbK complex in full dimensionality and report numerically exact quantum-mechanical reaction dynamics. The distribution of rotationally resolved rates is shown to be Poissonian. An analysis of the hyperspherical adiabatic potential curves explains this statistical character revealing a chaotic distribution for the short-range collision complex that plays a key role in governing the reaction outcome. We compare this with a lighter system with a smaller density of states (here the LiYbLi trimer) which displays random, and not chaotic, behavior.
Quantum reactive scattering calculations are reported for the ultracold hydrogen-exchange reaction and its non-reactive atom-exchange isotopic counterparts, proceeding from excited rotational states. It is shown that while the geometric phase (GP) does not necessarily control the reaction to all final states one can always find final states where it does. For the isotopic counterpart reactions these states can be used to make a measurement of the GP effect by separately measuring the even and odd symmetry contributions, which experimentally requires nuclear-spin final-state resolution. This follows from symmetry considerations that make the even and odd identical-particle exchange symmetry wavefunctions which include the GP locally equivalent to the opposite symmetry wavefunctions which do not. This equivalence reflects the important role discrete symmetries play in ultracold chemistry generally and highlights the key role ultracold reactions can play in understanding fundamental aspects of chemical reactivity.
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.
A five-dimensional coupled states (5D-CS) approximation is used to compute cross sections and rate coefficients for CO+H2 collisions. The 5D-CS calculations are benchmarked against accurate six-dimensional close-coupling (6D-CC) calculations for transitions between low-lying rovibrational states. Good agreement between the two formulations is found for collision energies greater than 10 cm-1. The 5D-CS approximation is then used to compute two separate databases which include highly excited states of CO that are beyond the practical limitations of the 6D-CC method. The first database assumes an internally frozen H2 molecule and allows rovibrational transitions for v < 5 and j < 30, where v and j are the vibrational and rotational quantum numbers of the initial state of the CO molecule. The second database allows H2 rotational transitions for initial CO states with v < 5 and j < 10. The two databases are in good agreement with each other for transitions that are common to both basis sets. Together they provide data for astrophysical models which were previously unavailable.
This paper deals with a class of nonparametric two-sample tests for ordered alternatives. The test statistics proposed are based on the number of observations from one sample that precede or exceed a threshold specified by the other sample, and they are extensions of Šidák's test. We derive their exact null distributions and also discuss a large-sample approximation. We then study their power properties exactly against the Lehmann alternative and make some comparative comments. Finally, we present an example to illustrate the proposed tests.
The results of accurate quantum reactive scattering calculations for the D + HD($v=4$, $j=0$) $\to$ D + HD($v'$, $j'$), D + HD($v=4$, $j=0$) $\to$ H + D$_2$($v'$, $j'$) and H + D$_2$($v=4$, $j=0$) $\to$ D + HD($v'$,$j'$) reactions are presented for collision energies between $1\,μ{\rm K}$ and $100\,{\rm K}$. The ${\it ab\ initio}$ BKMP2 PES for the ground electronic state of H$_3$ is used and all values of total angular momentum between $J=0-4$ are included. The general vector potential approach is used to include the geometric phase. The rotationally resolved, vibrationally resolved, and total reaction rate coefficients are reported as a function of collision energy. Rotationally resolved differential cross sections are also reported as a function of collision energy and scattering angle. Large geometric phase effects appear in the ultracold reaction rate coefficients which result in a significant enhancement or suppression of the rate coefficient (up to $3$ orders of magnitude) relative to calculations which ignore the geometric phase. The results are interpreted using a new quantum interference mechanism which is unique to ultracold collisions. Significant effects of the geometric phase also appear in the rotationally resolved differential cross sections which lead to a very different oscillatory structure in both energy and scattering angle. Several shape resonances occur in the $1$ - $10\,{\rm K}$ energy range and the geometric phase is shown to significantly alter the predicted resonance spectrum. The geometric phase effects depend sensitively on the nuclear spin which may provide experimentalists with the ability to control the reaction by the selection of a particular nuclear spin state.
We present calculated rate coefficients for ro-vibrational transitions of CO in collisions with H atoms for a gas temperature range of 10 K $\leq T \leq$ 3000 K, based on the recent three-dimensional ab initio H-CO interaction potential of Song et al(2013). Rate coefficients for ro-vibrational $v=1,j=0-30 \rightarrow v'=0, j'$ transitions were obtained from scattering cross sections previously computed with the close-coupling method by Song et al(2015). Combining these with the rate coefficients for vibrational $v=1-5 \rightarrow v' < v$ quenching obtained with the infinite-order sudden approximation, we propose a new extrapolation scheme that yields the rate coefficients for ro-vibrational $v=2-5,j=0-30 \rightarrow v',j'$ de-excitation. Cross sections and rate coefficients for ro-vibrational $v=2, j=0-30 \rightarrow v'=1,j'$ transitions calculated with the close-coupling method confirm the effectiveness of this extrapolation scheme. Our calculated and extrapolated rates are very different from those that have been adopted in the modeling of many astrophysical environments. The current work provides the most comprehensive and accurate set of ro-vibrational de-excitation rate coefficients for the astrophysical modeling of the H-CO collision system. Application of the previously available and new data sets in astrophysical slab models shows that the line fluxes typically change by 20-70% in high temperature environments (800 K) with an H/H$_2$ ratio of 1; larger changes occur for lower temperatures.
Carbon monoxide is a simple molecule present in many astrophysical environments, and collisional excitation rate coefficients due to the dominant collision partners are necessary to accurately predict spectral line intensities and extract astrophysical parameters. We report new quantum scattering calculations for rotational deexcitation transitions of CO induced by H using the three-dimensional potential energy surface~(PES) of Song et al. (2015). State-to-state cross sections for collision energies from 10$^{-5}$ to 15,000~cm$^{-1}$ and rate coefficients for temperatures ranging from 1 to 3000~K are obtained for CO($v=0$, $j$) deexcitation from $j=1-45$ to all lower $j'$ levels, where $j$ is the rotational quantum number. Close-coupling and coupled-states calculations were performed in full-dimension for $j$=1-5, 10, 15, 20, 25, 30, 35, 40, and 45 while scaling approaches were used to estimate rate coefficients for all other intermediate rotational states. The current rate coefficients are compared with previous scattering results using earlier PESs. Astrophysical applications of the current results are briefly discussed.