Estimation of entropy and other functionals of a multivariate density

For a multivariate density f with respect to Lebesgue measure , the estimation of % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% Waa8qaaeaacaWGkbGaaiikaiaadAgacaGGPaGaamOzaiaadsgacqaH% 8oqBaSqabeqaniabgUIiYdaaaa!4404!\[\int {J(f)fd\mu } \], and in particular % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% Waa8qaaeaacaWGMbWaaWbaaSqabeaacaaIYaaaaOGaamizaiabeY7a% TbWcbeqab0Gaey4kIipaaaa!41E4!\[\int {f^2 d\mu } \] and % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% Waa8qaaeaacaWGMbGaciiBaiaac+gacaGGNbGaamOzaiaadsgacqaH% 8oqBaSqabeqaniabgUIiYdaaaa!44AC!\[\int {f\log fd\mu } \], is studied. These two particular functionals are important in a number of contexts. Asymptotic bias and variance terms are obtained for the estimators % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% WaaybyaeqaleqabaGaey4jIKnaneaacaWGjbaaaOGaeyypa0Zaa8qa% aeaacaWGkbGaaiikamaawagabeWcbeqaaiabgEIizdqdbaGaamOzaa% aakiaacMcacaWGKbGaamOramaaBaaaleaacaWGobaabeaaaeqabeqd% cqGHRiI8aaaa!4994!\[\mathop I\limits^ \wedge = \int {J(\mathop f\limits^ \wedge )dF_N } \] and % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% WaaybyaeqaleqabaGaeSipIOdaneaacaWGjbaaaOGaeyypa0Zaa8qa% aeaacaWGkbGaaiikamaawagabeWcbeqaaiabgEIizdqdbaGaamOzaa% aakiaacMcadaGfGbqabSqabeaacqGHNis2a0qaaiaadAgaaaGccaWG% KbGaeqiVd0galeqabeqdcqGHRiI8aaaa!4C40!\[\mathop I\limits^ \sim = \int {J(\mathop f\limits^ \wedge )\mathop f\limits^ \wedge d\mu } \], where % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% WaaybyaeqaleqabaGaey4jIKnaneaacaWGMbaaaaaa!3E9C!\[{\mathop f\limits^ \wedge }\] is a kernel density estimate of f and F _n is the empirical distribution function based on the random sample X ₁ ,..., X _n from f. For the two functionalsmentioned above, a first order bias term for Î can be made zero by appropriate choices of non-unimodal kernels. Suggestions for the choice of bandwidth are given; for % MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% WaaybyaeqaleqabaGaey4jIKnaneaacaWGjbaaaOGaeyypa0Zaa8qa% aeaadaGfGbqabSqabeaacqGHNis2a0qaaiaadAgaaaGccaWGKbGaam% OramaaBaaaleaacaWGobaabeaaaeqabeqdcqGHRiI8aaaa!476C!\[\mathop I\limits^ \wedge = \int {\mathop f\limits^ \wedge dF_N } \], a study of optimal bandwidth is possible.This research was supported by an NSERC Grant and a UBC Killam Research Fellowship.     

Likelihood estimation of soft-core interaction potentials for Gibbsian point patterns

The likelihood method is developed for the analysis of socalled regular point patterns. Approximating the normalizing factor of Gibbs canonical distribution, we simultaneously estimate two parameters, one for the scale and the other which measures the softness (or hardness), of repulsive interactions between points. The approximations are useful up to a considerably high density. Some real data are analyzed to illustrate the utility of the parameters for characterizing the regular point pattern.     

HCN laser scattering on the JIPP T-IIU tokamak

A 4-channel HCN laser scattering system has been developed and applied to the JIPP T-IIU tokamak. Main aim of this system is to study microturbulence excited in the plasma, especially in the high power ICRF heating experiment. The observed density fluctuations are in a frequency range of the electron diamagnetic drift wave and have broad frequency spectra. The main part of the wavenumber is found to satisfy the condition of . In the heating experiments, deterioration of the energy confinement time is observed, and the density fluctuation level increases with the increase of the heating power.     

Freezing temperatures and densities of solutions of some hydrocarbons and sodium oleate inN-methylacetamide. Evidence for a solvophobic interaction

The freezing temperatures and densities (at 31°C) of solutions of octane, nonane, decane, 3,3-diethylpentane, and sodium oleate inN-methylacetamide (NMA) have been measured. The molality of the freezing solution was calculated from the density. The solubilities of octane, nonane, and decane inN-methylacetamide are also reported. Apparent molal volumes calculated from the densities are close to the values in the pure hydrocarbons and are not strong functions of the concentration. This indicates the absence of any unusual packing effect. The calculated free energies of transfer of the hydrocarbons from pure hydrocarbon to NMA solution are much less negative than the corresponding values for water, showing that the bulk solvophobic interaction inN-methylacetamide is smaller than in water. This is consistent with the freezing temperatures of sodium oleate which show that micelles do not form below 0.1 mole-kg^–1. The osmotic coefficients of the hydrocarbons calculated from the freezing temperatures showed negative deviations from ideality that were larger for the hydrocarbons with the higher molecular weights. Two estimates of the pairwise solvophobic interaction inN-methylacetamide indicate that it is also smaller than in water. The solvophobic effect in this solvent does not include the large entropy and enthalpy effects found in aqueous solutions.     

Microheterogeneity in aqueous-organic solutions: Heat capacities,volumes and expansibilities of some alcohols,aminoalcohol and tertiary amines in water

The heat capacities per unit volume and the densities of aqueous solutions of 2-propanol, neopentanol, tert-amylalcohol, 2-amino-2-methylpropanol, triethylamine and diethylmethylamine were measured, in many cases as a function of temperature, over the whole mole fraction or solubility range. Apparent and partial molal heat capacities, volumes and expansibilities were derived. The concentration dependence of these functions suggest the existence of transitions in some of these systems, in the water-rich region, qualitatively similar to micellization. The large relaxation contribution observed with some of the thermodynamic functions of hydrophobic alcohols and amines suggests a reinforcement of hydrophobic hydration due to strong hydrogen-bonding interactions of the polar groups with water.     

Synthesis,molecular and crystal structures,and characteristic features of electronic structures of salicylamide-based B,Si-containing chelates

The reaction of 4-(2-hydroxybenzoyl)-2,2,6,6-tetramethyl-2,6-disilamorpholine with BF₃Et₂O afforded (O-B)-chelate 4-[2-(difluoroboroxy)benzoyl]-2,2,6,6-tetramethyl-2,6-disilamorpholine. Treatment of the latter with BF₃Et₂O or SOCl₂ gave rise to products of the disilamorpholine ring opening, viz., (O-B) chelate 2-(difluoroboroxy)-N,N-bis(dimethylfluorosilylmethyl)benzamide or 2-(difluoroboroxy)-N, N-bis(dimethylchlorosilylmethyl)benzamide, respectively. The structures of the compounds synthesized were confirmed by X-ray diffraction analysis and ¹H, ¹³C, and ²⁹Si NMR spectroscopy. High-precision X-ray diffraction study and quantum-chemical calculations demonstrated that the coordination OSi bond is absent in the two last-mentioned compounds.Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 9, pp. 1846–1853, September, 2004.     

Do intramolecular halogen bonds exist? Ab initio calculations and crystal structures’ evidences

Cl···O intramolecular contacts are analysed here. Hence the Cambridge Structural Database (CSD) search was performed to find five-membered Cl–C=C–C=O pseudo-rings closed through Cl···O intramolecular contacts. Such rings were analysed and it was found that some of Cl···O distances are smaller than the corresponding sum of van der Waals radii. Additionally ab initio MP2/6-311++G(d,p) calculations and AIM analyses were carried out here on 3-chloropropenal and its fluoro derivatives to deepen the nature of intramolecular Cl···O interactions, some of findings may indicate that they are attractive and stabilizing ones. This paper is dedicated to Prof. T. M. Krygowski on the occasion of his 70th birthday     

Some aspects of diffusion: fluctuations in reaction diffusion,and geometric influences in nonuniform media

Summary We deal with two diffusion problems: Space-integrated conserved entities characterizing very fast - diffusion - controlled reactions, such as time lags, etc. are universal. They are given by relationships which do not reflect the failure of the mean field hydrodynamic equations. We present another application which does not reflect this failure, for determining the surface flux via a diffusion controlled reaction producing a colored product. Another anomalous diffusion process we considered is transport through cellular materials whose cell sizes are highly nonuniform. We have analyzed the effects of extreme nonuniformity by considering fractal-like models of cellular solids. The diffusion current through these models can exhibit anomalous time-dependencies which are not predicted by the diffusion equation. In particular, it is shown that the initial diffusion current can be characterized by a power-law dependence on the time. Furthermore, the exponent of the power law is given in terms of the distribution of cell sizes in the fractal-like cellular solid.