On improving the generalization of SVM classifier

The generalization performance of the SVM classifier depends mainly on the VC dimension and the dimensionality of the data. By reducing the VC dimension of the SVM classifier, its generalization performance is expected to increase. In the present paper, we argue that the VC dimension of SVM classifier can be reduced by applying bootstrapping and dimensionality reduction techniques. Experimental results showed that bootstrapping the original data and bootstrapping the projected (dimensionally reduced) data improved the performance of the SVM classifier.

Generalization of deviated linear cyclic pursuit

Earlier work on cyclic pursuit systems has shown that using heterogeneous gains for agents in linear cyclic pursuit, the point of convergence (rendezvous point) can be chosen arbitrarily. But there are some restrictions on this set of reachable points. The use of deviated cyclic pursuit, as discussed in this paper, expands this set of reachable points to include points which are not reachable by any known linear cyclic pursuit scheme. The limits on the deviations are determined by stability considerations. Such limits have been analytically obtained in this paper along with results on the expansion in reachable set and the latter has also been verified through simulations.

Global response sensitivity analysis using probability distance measures and generalization of Sobol's analysis

The study introduces two new alternatives for global response sensitivity analysis based on the application of the L-2-norm and Hellinger's metric for measuring distance between two probabilistic models. Both the procedures are shown to be capable of treating dependent non-Gaussian random variable models for the input variables. The sensitivity indices obtained based on the L2-norm involve second order moments of the response, and, when applied for the case of independent and identically distributed sequence of input random variables, it is shown to be related to the classical Sobol's response sensitivity indices. The analysis based on Hellinger's metric addresses variability across entire range or segments of the response probability density function. The measure is shown to be conceptually a more satisfying alternative to the Kullback-Leibler divergence based analysis which has been reported in the existing literature. Other issues addressed in the study cover Monte Carlo simulation based methods for computing the sensitivity indices and sensitivity analysis with respect to grouped variables. Illustrative examples consist of studies on global sensitivity analysis of natural frequencies of a random multi-degree of freedom system, response of a nonlinear frame, and safety margin associated with a nonlinear performance function. (C) 2015 Elsevier Ltd. All rights reserved.

A Generalization of Gravity

I consider theories of gravity built not just from the metric and affine connection, but also other (possibly higher rank) symmetric tensor(s). The Lagrangian densities are scalars built from them, and the volume forms are related to Cayley's hyperdeterminants. The resulting diff-invariant actions give rise to geometric theories that go beyond the metric paradigm (even metric-less theories are possible), and contain Einstein gravity as a special case. Examples contain theories with generalizeations of Riemannian geometry. The 0-tensor case is related to dilaton gravity. These theories can give rise to new types of spontaneous Lorentz breaking and might be relevant for ``dark'' sector cosmology.

Generalization of Response Number for Dynamic Plastic Response of Shells Subjected to impulsive Loading

A dimensionless number, termed as response number in Zhao [Archive of Applied Mechanics 68 (1998) 524], has been suggested for the dynamic plastic response of beams and plates made up of rigidly perfect plastic materials subjected to dynamic loading. Many theoretical and experimental results can be reformulated into new concise forms with the response number. The concept of a new dimensionless number, response number, termed as Rn(n), is generalized in Zhao [Forschung im Ingenieurwesen 65 (1999) 107] to study the elastic, plastic, dynamic elastic as well as dynamic plastic buckling problems of columns, plates as well as shells. The response number Rn(n) is generalized to the dynamic behaviour of shells of various shapes in the present paper.

Consumer Expertise or Credit Risk? An empirical analysis of mortgage pricing

Loan mortgage interest rates are usually the result of a bank-customer negotiation process. Credit risk, consumer cross-buying potential, bundling, financial market competition and other features affecting the bargaining power of the parties could affect price. We argue that, since mortgage loan is a complex product, consumer expertise could be a relevant factor for mortgage pricing. Using data on mortgage loan prices for a sample of 1055 households for the year 2005 (Bank of Spain Survey of Household Finances, EFF-2005), and including credit risk, costs, potential capacity of the consumer to generate future business and bank competition variables, the regression results indicate that consumer expertise-related metrics are highly significant as predictors of mortgage loan prices. Other factors such as credit risk and consumer cross-buying potential do not have such a significant impact on mortgage prices. Our empirical results are affected by the credit conditions prior to the financial crisis and could shed some light on this issue.

Generalization of the JTZ model to open plane wakes

The JTZ model [C. Jung, T. T¶el and E. Ziemniak, Chaos 3, (1993) 555], as a theoretical model of a plane wake behind a circular cylinder in a narrow channel at a moderate Reynolds number, has previously been employed to analyze phenomena of chaotic scattering. It is ex- tended here to describe an open plane wake without the con¯ned nar- row channel by incorporating a double row of shedding vortices into the intermediate and far wake. The extended JTZ model is found in qualitative agreement with both direct numerical simulations and ex- perimental results in describing streamlines and vorticity contours. To further validate its applications to particle transport processes, the in- teraction between small spherical particles and vortices in an extended JTZ model °ow is studied. It is shown that the particle size has signif- icant in°uences on the features of particle trajectories, which have two characteristic patterns: one is rotating around the vortex centers and the other accumulating in the exterior of vortices. Numerical results based on the extended JTZ model are found in qualitative agreement with experimental ones in the normal range of particle sizes.

Credit financing of aquaculture programmes -- a review

A discussion is presented on the role of credit financing in the development of aquaculture programmes in Nigeria. The constraints militating against credit availability to aquaculture vis-a-vis the competition between aquaculture and other agricultural and industrial sectors for the allocation of available credit facilities are examined. Possible ways of enhancing preferential allocation of capital to the aquaculture industry from the various sources of credit available to the Nigerial agricultural economy are suggested

Projections in a normed linear space and a generalization of the pseudo-inverse

The concept of a "projection function" in a finite-dimensional real or complex normed linear space H (the function P_M which carries every element into the closest element of a given subspace M) is set forth and examined.

If dim M = dim H - 1, then P_M is linear. If P_N is linear for all k-dimensional subspaces N, where 1 ≤ k < dim M, then P_M is linear.

The projective bound Q, defined to be the supremum of the operator norm of P_M for all subspaces, is in the range 1 ≤ Q < 2, and these limits are the best possible. For norms with Q = 1, P_M is always linear, and a characterization of those norms is given.

If H also has an inner product (defined independently of the norm), so that a dual norm can be defined, then when P_M is linear its adjoint P_M^H is the projection on (kernel P_M)^⊥ by the dual norm. The projective bounds of a norm and its dual are equal.

The notion of a pseudo-inverse F⁺ of a linear transformation F is extended to non-Euclidean norms. The distance from F to the set of linear transformations G of lower rank (in the sense of the operator norm ∥F - G∥) is c/∥F⁺∥, where c = 1 if the range of F fills its space, and 1 ≤ c < Q otherwise. The norms on both domain and range spaces have Q = 1 if and only if (F⁺)⁺ = F for every F. This condition is also sufficient to prove that we have (F⁺)^H = (F^H)⁺, where the latter pseudo-inverse is taken using dual norms.

In all results, the real and complex cases are handled in a completely parallel fashion.

O credit scoring em apoio ao risco de crédito: a identificação de variáveis que contribuem no sucesso da implantação de empresas franqueadas

A concessão de crédito a empresas que participam do mercado consiste na entrega de um ativo em determinado momento, com a promessa de pagamento deste bem ou direito em data futura. Tal situação se configura como um evento incerto, pois existe a possibilidade de que tal obrigação não seja honrada pela promitente compradora, originando desta forma, o risco de crédito. Cabe à parte concessora do ativo que origina o risco de crédito, verificar a capacidade de seu cliente em cumprir o compromisso futuro assumido, analisando as variáveis que sugerem o sucesso da operação de crédito. As empresas que se encontram em fase de implantação caracterizam-se não somente pela ausência de histórico das variáveis acima, como também pelo aumento considerável do risco de continuidade. Tal situação é comprovada por pesquisas realizadas em empresas com até cinco anos de atuação. A impossibilidade na mensuração da capacidade de crédito proporcionada por este cenário, ocasiona severa restrição creditícia às empresas novas, principalmente ao crédito de longo prazo, imprescindível nesta fase de investimentos. Entretanto, esta restrição não se verifica em empresas de franquia, cujo empreendedor tem o privilégio de iniciar seu negócio com linhas de crédito de investimentos já prontas no mercado com esta finalidade. Este estudo objetiva identificar quais as características presentes em empresas franqueadas que permitem a concessão de crédito segura na fase de implantação por parte das instituições financeiras e se tais características podem discriminar variáveis que são determinantes no sucesso da franqueada proponente ao crédito bancário. A aplicação de análise fatorial em banco de dados com empresas de franquia permitiu identificar com sucesso um grupo de sete principais variáveis principais, que serviram de base a um modelo de regressão logística e análise discriminante. O modelo de regressão logística mostrou-se bom para a melhora da probabilidade de acerto de empresas solventes ao passo que a análise discriminante não apresentou melhora nesses resultados.

Safety net or debt net?: ties between marketing and credit in coastal Bangladesh

Three fishing communities in coastal Bangladesh namely: 1) Hindu community near Chittagong; 2) Muslim community in Rehania; and 3) Bhuddist community of Rakhaipara were studied, and each shared the same credit marketing system.