236 research outputs found
Noise inference for ergodic Lévy driven SDE
No full textWe study inference for the driving Lévy noise of an ergodic stochastic differential equation (SDE) model, when the process is observed at high-frequency and long time and when the drift and scale coefficients contain finite-dimensional unknown parameters. By making use of the Gaussian quasi-likelihood function for the coefficients, we derive a stochastic expansion for functionals of the unit-time residuals, which clarifies some quantitative effect of plugging-in the estimators of the coefficients, thereby enabling us to take several inference procedures for the driving-noise characteristics into account. We also present new classes and methods available in YUIMA for the simulation and the estimation of a Lévy SDE model. We highlight the flexibility of these new advances in YUIMA using simulated and real data
Quasi-likelihood analysis for Student-Lévy regression
No full textWe consider the quasi-likelihood analysis for a linear regression model driven by a Student-t Lévy process with constant scale and arbitrary degrees of freedom. The model is observed at high frequency over an extending period, under which we can quantify how the sampling frequency affects estimation accuracy. In that setting, joint estimation of trend, scale, and degrees of freedom is a non-trivial problem. The bottleneck is that the Student-t distribution is not closed under convolution, making it difficult to estimate all the parameters fully based on the high-frequency time scale. To efficiently deal with the intricate nature from both theoretical and computational points of view, we propose a two-step quasi-likelihood analysis: first, we make use of the Cauchy quasi-likelihood for estimating the regression-coefficient vector and the scale parameter; then, we construct the sequence of the unit-period cumulative residuals to estimate the remaining degrees of freedom. In particular, using full data in the first step causes a problem stemming from the small-time Cauchy approximation, showing the need for data thinning
Phenomenological description of technical resources applied to two works by Franz Liszt and two works by Hiromi Uehara
No full textEl presente trabajo estudia la aplicación de recursos técnicos en dos obras de Franz Liszt y dos de Hiromi Uehara. Se establecen objetivos claros que incluyen la descripción de los recursos técnicos y el sonido deseado para cada obra. La metodología se centra en el análisis físico de la ejecución musical y la interacción con el teclado. A través de un enfoque práctico, el autor pretende ofrecer una perspectiva sobre la interpretación de estos compositores, así como herramientas útiles para otros pianistas, contribuyendo a su formación técnica y artística.The present work studies the application of technical resources in two works by Franz Liszt and two by Hiromi Uehara. Clear objectives are established, including the description of the technical resources and the desired sound for each work. The methodology focuses on the physical analysis of the musical performance and the interaction with the keyboard. Through a practical approach, the author aims to offer a perspective on the interpretation of these composers, as well as useful tools for other pianists, contributing to their technical and artistic training
Statistical inference for misspecified ergodic Lévy driven stochastic differential equation models
No full text
青年期の朗読教育
No full textIn Uehara 2001, the author discussed significance and theoretical background of reading education, practical issues and challenges. The author also proposed a reading education that fuses the art and science. Since then the author has explored such education while conducting basic research and practical education of “Reading to Freedom” as a systematic language education. Now the author would like to report a thought on education and make it as a starting point of a new research and education. This article discusses reading education in adolescence from the perspective of education of words as a place of art training.departmental bulletin pape
「自由への読書」のための基礎的研究3－平和主義的感性の育成－
No full textThe author discussed in Uehara 2007 and Uehara 2009 the importance of improving three abilities－ "imagination", "critical thinking", and "meta-cognitive ability that supports selfcontrol" －in order to lead the rising generation to the state of "living freely" as the goal of education. The author also argued that reading is the best method to improve the three abilities, advocated "Reading for Freedom" --a systematic reading education program that meets the needs of children's stage of development, and made practical reports. "Reading for Freedom" is an educational attempt to develop a self-disciplined, truly free human being. The author hopes for a sustainable society created by truly free humans where all lives can and will be able to coexist. In this troubled times, the source of hope for their future should exist in an environment where the goal of "cultivation of a sense of pacifism" should lie as thorough bass at the bottom of various forms of education. In this article, the author restructures her comprehension that was acquired through her lectures and workshops where she learned practical lessons with various people who are involved in education. Based on a belief that a "systematic reading education program" is a method of "cultivation of a sense of pacifism", she discusses the significance of reading education and the responsibility of educators.departmental bulletin pape
誤特定非正規確率微分方程式モデルの推論
No full textOpen House, ISM in National Center of Sciences Building, 2019.6.05統計数理研究所オープンハウス（学術総合センター）、R1.6.5ポスター発
Quasi-Likelihood Analysis for Student-L\'evy Regression
No full textWe consider the quasi-likelihood analysis for a linear regression model
driven by a Student-t L\'{e}vy process with constant scale and arbitrary
degrees of freedom. The model is observed at high frequency over an extending
period, under which we can quantify how the sampling frequency affects
estimation accuracy. In that setting, joint estimation of trend, scale, and
degrees of freedom is a non-trivial problem. The bottleneck is that the
Student-t distribution is not closed under convolution, making it difficult to
estimate all the parameters fully based on the high-frequency time scale. To
efficiently deal with the intricate nature from both theoretical and
computational points of view, we propose a two-step quasi-likelihood analysis:
first, we make use of the Cauchy quasi-likelihood for estimating the
regression-coefficient vector and the scale parameter; then, we construct the
sequence of the unit-period cumulative residuals to estimate the remaining
degrees of freedom. In particular, using full data in the first step causes a
problem stemming from the small-time Cauchy approximation, showing the need for
data thinning
- …
