This article comprehensively reviews some applications of Hawkes process to credit risk modeling with “contagion effect”. Credit risk is the risk associated with financial losses caused by credit events such as debtorsʼ defaults or credit rating transitions. Financial institutions are required to assess more accurately total credit risk of their large credit portfolios for better risk managements. As such, credit risk quantification models are desired to capture the effect of credit risk contagions, which may cause extreme financial losses. Hawkes process is a nonnegative integer-valued stochastic process which has been often used as a basic model for counting contagious events such as infectious diseases in epidemiology and earthquake in seismology. Similarly, modeling with Hawkes process enables us to easily capture some features of contagious credit events and thus to improve the performance of assessing total credit risks. In addition, a multivariate Hawkes process has capability of estimating mutual contagion effects among different industrial sectors. In this article, as for credit risk modeling and analyses with Hawkes processes, not only an introductory theoretical review but some illustrative results from some recent works of the present authors of empirical analyses are presented.
In order to solve the problems among community-dwelling elderly persons, mutual support is indispensable. In this paper, an exercise program in an active aging project conducted by the authors is introduced. The aim of this project is the promotion of mutual support among the communitydwelling persons. Regarding this program, there arises a mathematical problem, that is, evaluation of the results. Although the interactions between the participants can be measured by a wearable sensor, it is difficult to theoretically evaluate from the data if the significant change is brought about by the program. In this paper, a statistical hypothesis testing for this evaluation is presented. In the test, a network model is used as a statistical model of the networks. Application to testing gender difference is also shown.
This paper deals with Riemannian optimization, that is, optimization on Riemannian manifolds. Theories of Euclidean optimization and Riemannian manifolds are first briefly reviewed together with some simple and motivating examples, followed by the Riemannian optimization theory. Retractions and vector transports on Riemannian manifolds are introduced according to the literature to describe a general Riemannian optimization algorithm. Recent convergence analysis results of several types of Riemannian conjugate gradient methods, such as Fletcher-Reeves and Dai-Yuan-types, are then given and discussed in detail. Some applications of Riemannian optimization to problems of current interest, such as 1)singular value decomposition in numerical linear algebra; 2)canonical correlation analysis and topographic independent component analysis as statistical methods; 3)low-rank tensor completion for machine learning; 4)optimal model reduction in control theory; and 5)doubly stochastic inverse eigenvalue problem, are also introduced.