Régression logistique bayésienne : comparaison de densités a priori

La régression logistique est un modèle de régression linéaire généralisée (GLM) utilisé pour des variables à expliquer binaires. Le modèle cherche à estimer la probabilité de succès de cette variable par la linéarisation de variables explicatives. Lorsque l’objectif est d’estimer le plus précisément l’impact de différents incitatifs d’une campagne marketing (coefficients de la régression logistique), l’identification de la méthode d’estimation la plus précise est recherchée. Nous comparons, avec la méthode MCMC d’échantillonnage par tranche, différentes densités a priori spécifiées selon différents types de densités, paramètres de centralité et paramètres d’échelle. Ces comparaisons sont appliquées sur des échantillons de différentes tailles et générées par différentes probabilités de succès. L’estimateur du maximum de vraisemblance, la méthode de Gelman et celle de Genkin viennent compléter le comparatif. Nos résultats démontrent que trois méthodes d’estimations obtiennent des estimations qui sont globalement plus précises pour les coefficients de la régression logistique : la méthode MCMC d’échantillonnage par tranche avec une densité a priori normale centrée en 0 de variance 3,125, la méthode MCMC d’échantillonnage par tranche avec une densité Student à 3 degrés de liberté aussi centrée en 0 de variance 3,125 ainsi que la méthode de Gelman avec une densité Cauchy centrée en 0 de paramètre d’échelle 2,5.

Variabilité des souches de Escherichia coli provenant de divers poulaillers au Québec

La technique d’empreinte génétique par rep-PCR, qui utilise des séquences d’ADN répétitives, a été utilisée pour mettre en évidence la présence de groupes d’Escherichia coli signatures pour divers poulaillers et d’évaluer leur évolution suite au détassement. L’amorce (GTG)5 a été utilisée pour générer des empreintes d’ADN de 522 isolats provenant de 7 poulaillers échantillonnés deux fois : juste avant et 5 jours après le détassement. Les empreintes d’ADN ont été analysées selon l’algorithme de correspondance de bandes de Jaccard. Les analyses de Jackknife des coefficients de similitude ont révélé qu’entre 73% et 93% des isolats ont pu être correctement regroupés selon leur poulailler d’origine. Un dendrogramme construit à partir des coefficients de similitude de Jaccard a groupé les isolats dans 42 grappes avec près de la moitié dans une seule grappe. Environ 80% des isolats ont été groupés dans les 6 plus grosses grappes. Quatre de ces grappes été constituées majoritairement d’isolats provenant d’un seul site. Ces grappes pourraient être des grappes signatures qui permettraient d’identifier des poulaillers en particulier. La comparaison des nombres de grappes présentes avant et après le détassement a révélé une variabilité de l’impact du détassement sur les populations fécales d’E. coli. Pour certains sites, il y avait peu d’agrégats présents tant avant qu’après le détassement alors que pour d’autres sites c’était le contraire. Quoique plus de recherches soient nécessaires afin de valider les conclusions, nos résultats suggèrent la présence de sous-populations signatures d’E. coli pour certains poulaillers et une réponse variable à l’effet du détassement.

Coefficients de Clebsch-Gordan de la super-algèbre osp(1|2)

Les fonctions génératrices des coefficients de Clebsch Gordan pour la superalgèbre de Lie osp(1|2) sont dérivées en utilisant deux approches. Une première approche généralise une méthode proposée par Granovskii et Zhedanov pour l'appliquer dans le cas de osp(1|2), une algèbre dont le coproduit est torsadé. Une seconde approche repose sur la réalisation de osp(1|2) en tant qu'algèbre dynamique d'un oscillateur parabosonique et utilise une équivalence dans cette réalisation entre le changements de coordonnées polaires à cartésiennes et le problème de Clebsch-Gordan. Un chapitre moins formel précède ces dérivations et présente comment le problème de Clebsch-Gordan s'interprète en tant que réalisation d'une algèbre de fusion. La notion abstraite de fusion est introduite, soulignant son importance en physique, pour en venir au cas particulier du problème de Clebsch-Gordan. Un survol du cas de l'algèbre osp(1|2) et de ses utilisations en physique mathématique conclut ce chapitre.

Effect of polymer architecture on Curcumin 1 encapsulation and release from PEGylated polymer nanoparticles: toward a drug delivery nano-platform to the CNS

We developed a nanoparticles (NPs) library from poly(ethylene glycol)–poly lactic acid comb-like polymers with variable amount of PEG. Curcumin was encapsulated in the NPs with a view to develop a delivery platform to treat diseases involving oxidative stress affecting the CNS. We observed a sharp decrease in size between 15 and 20% w/w of PEG which corresponds to a transition from a large solid particle structure to a “micelle-like” or “polymer nano-aggregate” structure. Drug loading, loading efficacy and release kinetics were determined. The diffusion coefficients of curcumin in NPs were determined using a mathematical modeling. The higher diffusion was observed for solid particles compared to “polymer nano-aggregate” particles. NPs did not present any significant toxicity when tested in vitro on a neuronal cell line. Moreover, the ability of NPs carrying curcumin to prevent oxidative stress was evidenced and linked to polymer architecture and NPs organization. Our study showed the intimate relationship between the polymer architecture and the biophysical properties of the resulting NPs and sheds light on new approaches to design efficient NP-based drug carriers.

A multiple-building diffraction attenuation function expressed in terms of UTD coefficients for microcellular communications

This work presents an explicit formulation for multiple- edge diffraction for mobile radiowave propagation in terms of uniform theory of diffraction (UTD) coefficients when a spherical incident wave is considered. This solution can be used in an UTD context and sharply reduces the computing time over existing formulation. Results can be applied in the planning of microcellular systems

Salting Coefficients of Fluorobenzoic Acids

The Setschenow parameter and thermodynamic parameters of transfer of 2-, 3- and 4-fluorobenzoic acid from water to salt solution are reported. The data have been rationalized by considering the structure breaking effects of the ions of the salts, the localized hydrolysis model and the internal pressure theory.

Salting Coefficients of 2-, 3-, and 4-Methylbenzoic Acids

The Setschenow parameter and thermodynamic parameters of transfer of 2-, 3-, and 4-methylbenzoic acids from water to salt solutions have been reported. The data have been rationalized by considering the structure breaking effects of the ions of the salts, the localized hydrolysis model, and the internal pressure theory.

An Improved Technique for Evolving Wavelet Coefficients for Fingerprint Image Compression

In this paper, an improved technique for evolving wavelet coefficients refined for compression and reconstruction of fingerprint images is presented. The FBI fingerprint compression standard [1, 2] uses the cdf 9/7 wavelet filter coefficients. Lifting scheme is an efficient way to represent classical wavelets with fewer filter coefficients [3, 4]. Here Genetic algorithm (GA) is used to evolve better lifting filter coefficients for cdf 9/7 wavelet to compress and reconstruct fingerprint images with better quality. Since the lifting filter coefficients are few in numbers compared to the corresponding classical wavelet filter coefficients, they are evolved at a faster rate using GA. A better reconstructed image quality in terms of Peak-Signal-to-Noise-Ratio (PSNR) is achieved with the best lifting filter coefficients evolved for a compression ratio 16:1. These evolved coefficients perform well for other compression ratios also.

Faster techniques to evolve wavelet coefficients for better fingerprint image compression

In this article, techniques have been presented for faster evolution of wavelet lifting coefficients for fingerprint image compression (FIC). In addition to increasing the computational speed by 81.35%, the coefficients performed much better than the reported coefficients in literature. Generally, full-size images are used for evolving wavelet coefficients, which is time consuming. To overcome this, in this work, wavelets were evolved with resized, cropped, resized-average and cropped-average images. On comparing the peak- signal-to-noise-ratios (PSNR) offered by the evolved wavelets, it was found that the cropped images excelled the resized images and is in par with the results reported till date. Wavelet lifting coefficients evolved from an average of four 256 256 centre-cropped images took less than 1/5th the evolution time reported in literature. It produced an improvement of 1.009 dB in average PSNR. Improvement in average PSNR was observed for other compression ratios (CR) and degraded images as well. The proposed technique gave better PSNR for various bit rates, with set partitioning in hierarchical trees (SPIHT) coder. These coefficients performed well with other fingerprint databases as well.

Evolution of Better Wavelet Coefficients for Fingerprint Image Compression using cropped images

This paper explains the Genetic Algorithm (GA) evolution of optimized wavelet that surpass the cdf9/7 wavelet for fingerprint compression and reconstruction. Optimized wavelets have already been evolved in previous works in the literature, but they are highly computationally complex and time consuming. Therefore, in this work, a simple approach is made to reduce the computational complexity of the evolution algorithm. A training image set comprised of three 32x32 size cropped images performed much better than the reported coefficients in literature. An average improvement of 1.0059 dB in PSNR above the classical cdf9/7 wavelet over the 80 fingerprint images was achieved. In addition, the computational speed was increased by 90.18 %. The evolved coefficients for compression ratio (CR) 16:1 yielded better average PSNR for other CRs also. Improvement in average PSNR was experienced for degraded and noisy images as well

Study on period search methods of variable stars: Application to ASAS and CRTS databases

Study on variable stars is an important topic of modern astrophysics. After the invention of powerful telescopes and high resolving powered CCD’s, the variable star data is accumulating in the order of peta-bytes. The huge amount of data need lot of automated methods as well as human experts. This thesis is devoted to the data analysis on variable star’s astronomical time series data and hence belong to the inter-disciplinary topic, Astrostatistics. For an observer on earth, stars that have a change in apparent brightness over time are called variable stars. The variation in brightness may be regular (periodic), quasi periodic (semi-periodic) or irregular manner (aperiodic) and are caused by various reasons. In some cases, the variation is due to some internal thermo-nuclear processes, which are generally known as intrinsic vari- ables and in some other cases, it is due to some external processes, like eclipse or rotation, which are known as extrinsic variables. Intrinsic variables can be further grouped into pulsating variables, eruptive variables and flare stars. Extrinsic variables are grouped into eclipsing binary stars and chromospheri- cal stars. Pulsating variables can again classified into Cepheid, RR Lyrae, RV Tauri, Delta Scuti, Mira etc. The eruptive or cataclysmic variables are novae, supernovae, etc., which rarely occurs and are not periodic phenomena. Most of the other variations are periodic in nature. Variable stars can be observed through many ways such as photometry, spectrophotometry and spectroscopy. The sequence of photometric observa- xiv tions on variable stars produces time series data, which contains time, magni- tude and error. The plot between variable star’s apparent magnitude and time are known as light curve. If the time series data is folded on a period, the plot between apparent magnitude and phase is known as phased light curve. The unique shape of phased light curve is a characteristic of each type of variable star. One way to identify the type of variable star and to classify them is by visually looking at the phased light curve by an expert. For last several years, automated algorithms are used to classify a group of variable stars, with the help of computers. Research on variable stars can be divided into different stages like observa- tion, data reduction, data analysis, modeling and classification. The modeling on variable stars helps to determine the short-term and long-term behaviour and to construct theoretical models (for eg:- Wilson-Devinney model for eclips- ing binaries) and to derive stellar properties like mass, radius, luminosity, tem- perature, internal and external structure, chemical composition and evolution. The classification requires the determination of the basic parameters like pe- riod, amplitude and phase and also some other derived parameters. Out of these, period is the most important parameter since the wrong periods can lead to sparse light curves and misleading information. Time series analysis is a method of applying mathematical and statistical tests to data, to quantify the variation, understand the nature of time-varying phenomena, to gain physical understanding of the system and to predict future behavior of the system. Astronomical time series usually suffer from unevenly spaced time instants, varying error conditions and possibility of big gaps. This is due to daily varying daylight and the weather conditions for ground based observations and observations from space may suffer from the impact of cosmic ray particles. Many large scale astronomical surveys such as MACHO, OGLE, EROS, xv ROTSE, PLANET, Hipparcos, MISAO, NSVS, ASAS, Pan-STARRS, Ke- pler,ESA, Gaia, LSST, CRTS provide variable star’s time series data, even though their primary intention is not variable star observation. Center for Astrostatistics, Pennsylvania State University is established to help the astro- nomical community with the aid of statistical tools for harvesting and analysing archival data. Most of these surveys releases the data to the public for further analysis. There exist many period search algorithms through astronomical time se- ries analysis, which can be classified into parametric (assume some underlying distribution for data) and non-parametric (do not assume any statistical model like Gaussian etc.,) methods. Many of the parametric methods are based on variations of discrete Fourier transforms like Generalised Lomb-Scargle peri- odogram (GLSP) by Zechmeister(2009), Significant Spectrum (SigSpec) by Reegen(2007) etc. Non-parametric methods include Phase Dispersion Minimi- sation (PDM) by Stellingwerf(1978) and Cubic spline method by Akerlof(1994) etc. Even though most of the methods can be brought under automation, any of the method stated above could not fully recover the true periods. The wrong detection of period can be due to several reasons such as power leakage to other frequencies which is due to finite total interval, finite sampling interval and finite amount of data. Another problem is aliasing, which is due to the influence of regular sampling. Also spurious periods appear due to long gaps and power flow to harmonic frequencies is an inherent problem of Fourier methods. Hence obtaining the exact period of variable star from it’s time series data is still a difficult problem, in case of huge databases, when subjected to automation. As Matthew Templeton, AAVSO, states “Variable star data analysis is not always straightforward; large-scale, automated analysis design is non-trivial”. Derekas et al. 2007, Deb et.al. 2010 states “The processing of xvi huge amount of data in these databases is quite challenging, even when looking at seemingly small issues such as period determination and classification”. It will be beneficial for the variable star astronomical community, if basic parameters, such as period, amplitude and phase are obtained more accurately, when huge time series databases are subjected to automation. In the present thesis work, the theories of four popular period search methods are studied, the strength and weakness of these methods are evaluated by applying it on two survey databases and finally a modified form of cubic spline method is intro- duced to confirm the exact period of variable star. For the classification of new variable stars discovered and entering them in the “General Catalogue of Vari- able Stars” or other databases like “Variable Star Index“, the characteristics of the variability has to be quantified in term of variable star parameters.

Duplication coefficients via generating functions

In this paper, we solve the duplication problem P_n(ax) = sum_{m=0}^{n}C_m(n,a)P_m(x) where {P_n}_{n>=0} belongs to a wide class of polynomials, including the classical orthogonal polynomials (Hermite, Laguerre, Jacobi) as well as the classical discrete orthogonal polynomials (Charlier, Meixner, Krawtchouk) for the specific case a = −1. We give closed-form expressions as well as recurrence relations satisfied by the duplication coefficients.

On the Computation of Fourier Coefficients

In this paper we derive an identity for the Fourier coefficients of a differentiable function f(t) in terms of the Fourier coefficients of its derivative f'(t). This yields an algorithm to compute the Fourier coefficients of f(t) whenever the Fourier coefficients of f'(t) are known, and vice versa. Furthermore this generates an iterative scheme for N times differentiable functions complementing the direct computation of Fourier coefficients via the defining integrals which can be also treated automatically in certain cases.

Fluoreszenz und Lasertätigkeit in dünnen amorphen Schichten von Spirobifluorenderivaten

Es wurden die optischen Eigenschaften einiger Spirooligophenylverbindungen untersucht. Bei den Verbindungen handelte es sich um lineare und verzweigte Derivate des Spirobifluorens. Es wurden dünne amorphe Schichten der Verbindungen mittels spincoating und Vakuumverdampfen (OMBD) auf unterschiedlichen Substraten hergestellt. Mit spektroskopischer Ellipsometrie konnten die Schichtdicken und optische Konstanten der dünnen Schichten bestimmt werden. Dafür sind die Extinktionsspektren mit Tauc-Lorentz-Oszillatoren modelliert worden. Auf diese Weise ließen sich die optischen Konstanten der amorphen Filme besser beschreiben als mit den üblicherweise verwendeten Gauß-Oszillatoren. In dünnen Filmen von Spirosexiphenyl konnte uniaxiale Anisotropie nachgewiesen werden. Im Bereich der pie,pie*-Bande beträgt das Verhältnis des ordentlichen zum außerordentlichen Extinktionskoeffizienten 2.09. Mit einer Integrationskugel wurden die absoluten Quantenausbeuten der Fluoreszenz in festen Filmen bestimmt. Dafür ist ein vereinfachtes Verfahren der Auswertung entwickelt worden. Im Vergleich der untersuchten Substanzen zeigt sich, dass die Verbindungen mit dem Sexiphenyl-Chromophor höhere Quantenausbeuten im Festkörper haben (31 % - 48 %) als die Verbindungen mit Quaterphenyl als Chromophor (15 % - 30 %). In den beiden Klassen haben jeweils die sterisch anspruchsvollen Octopusvarianten die höchsten Festkörperquantenausbeuten. Durch verdünnen mit m,m-Spirosexiphenyl konnte die Quantenausbeute von p,p-Spirosexiphenyl in dünnen festen Filmen bis auf 65 % (95 % m,m- und 5 % p,p-Spirosexiphenyl) gesteigert werden. Eine Korrelation der Quantenausbeuten in Lösung und im festen, unverdünnten Film wurde nicht festgestellt. Als dünne Filme in Wellenleitergeometrie zeigen Spirooligophenyle bei optischer Anregung verstärkte spontane Emission (ASE). Dies manifestiert sich in einer Einengung des Emissionsspektrums mit zunehmender Pumpleistungsdichte. Auch für stimulierte Emission sind die Verbindungen mit Sexiphenylchromophor besser geeignet. Die niedrigste Schwelle in einer unverdünnten Reinsubstanz wurde mit 0.23 µJ/cm² in einer aufgeschleuderten Schicht Spirosexiphenyl gemessen. Auch 4-Spiro³, Spiro-SPO und Octo-2 zeigten niedrige ASE-Schwellen von 0.45 µJ/cm², 0.45 µJ/cm² und 0.5 µJ/cm². Die ASE-Schwellwerte von Spiroquaterphenyl und seinen beiden Derivaten Methoxyspiroquaterphenyl und Octo-1 sind mit 1.8 µJ/cm², 1.4 µJ/cm² und 1.2 µJ/cm² höher als die der Sexiphenylderivate. Im gemischten System aus m,m- und p,p-Spirosexiphenyl konnte die ASE-Schwelle noch weiter gesenkt werden. Bei einer Konzentration von 5 % p,p-Spirosexiphenyl wurde ein Schwellwert von nur 100 nJ/cm² bestimmt. Durch Dotierung mit unterschiedlichen Farbstoffen in Spirosexiphenyl als Matrix konnte ASE fast über den gesamten sichtbaren Spektralbereich gezeigt werden. Mit der „variable Streifenlänge“ (VSL-) -Methode wurden die pumpleistungsabhängigen Gainspektren dünner aufgedampfter Proben gemessen. Hieraus konnten die Wechselwirkungsquerschnitte der stimulierten Emission der Substanzen ermittelt werden. In Übereinstimmung mit den Verhältnissen bei den Festkörperfluoreszenzquantenausbeuten und den Schwellwerten der ASE sind auch bei den Gainkoeffizienten reiner Spirooligophenyle die besten Werte bei den Sexiphenylderivaten gefunden worden. Der Wirkungsquerschnitt der stimulierten Emission beträgt für Methylspiroquaterphenyl und Octo-1 ca. 1.8*10^-17 cm². Für Spiro-SPO und Spirosexiphenyl wurden Wirkungsquerschnitte von 7.5*10^-17 cm² bzw. 9.2*10^-17 cm² bestimmt. Noch etwas größer waren die Werte im gemischten System aus m,m- und p,p-Spirosexiphenyl (1.1*10^-16 cm²) und für DPAVB dotiert in Spirosexiphenyl (1.4*10^-16 cm²). Der höchste Maximalwert des Gainkoeffizienten von 328 cm-1 bei einer absorbierten Pumpenergiedichte von 149 µJ/cm² wurde mit Spirosexiphenyl erreicht. Abschließend wurden DFB-Laser-Strukturen mit reinen und dotierten Spirooligophenylverbindungen als aktiven Materialien vorgestellt. Mit Spiroterphenyl konnte ein DFB-Laser mit der bisher kürzesten Emissionswellenlänge (361.9 nm) in einem organischen Festkörperlaser realisiert werden. Mit reinen Spirooligophenylverbindungen und Mischungen daraus habe ich DFB-Lasing bei Wellenlängen zwischen 361.9 nm und 479 nm aufgezeigt. Durch Dotierung mit DPAVB wurde der Bereich der erreichbaren Wellenlängen bis 536 nm erweitert, bei gleichzeitiger Erniedrigung der Schwellenergiedichten für Lasertätigkeit. Bei Emissionswellenlängen von 495 nm bis 536 nm blieb die Laserschwelle zwischen 0.8 µJ/cm² und 1.1 µJ/cm². Diese Werte sind für DFB-Laser zweiter Ordnung sehr niedrig und geben Anlass zu glauben, dass sich mit DFB-Strukturen erster Ordnung Schwellen im Nanojoule Bereich erzielen lassen. Damit würde man den Bedingungen für elektrisch gepumpten Betrieb nahe kommen.

On Solutions of Holonomic Divided-Difference Equations on Non-Uniform Lattices

The main aim of this paper is the development of suitable bases (replacing the power basis x^n (n\in\IN_\le 0) which enable the direct series representation of orthogonal polynomial systems on non-uniform lattices (quadratic lattices of a discrete or a q-discrete variable). We present two bases of this type, the first of which allows to write solutions of arbitrary divided-difference equations in terms of series representations extending results given in [16] for the q-case. Furthermore it enables the representation of the Stieltjes function which can be used to prove the equivalence between the Pearson equation for a given linear functional and the Riccati equation for the formal Stieltjes function. If the Askey-Wilson polynomials are written in terms of this basis, however, the coefficients turn out to be not q-hypergeometric. Therefore, we present a second basis, which shares several relevant properties with the first one. This basis enables to generate the defining representation of the Askey-Wilson polynomials directly from their divided-difference equation. For this purpose the divided-difference equation must be rewritten in terms of suitable divided-difference operators developed in [5], see also [6].