Academics

Simultaneous Inference for Eigensystems and FPC Scores of Functional Data

Time:Mon., 11:00-12:00 Dec. 16, 2024

Venue:C548, Shuangqing Complex Building A

Speaker:Qirui Hu

Statistical Seminar

Organizer:

吴宇楠

Speaker:

胡祺睿

清华大学统计学研究中心

Time:

Mon., 11:00-12:00

Dec. 16, 2024

Venue:

C548, Shuangqing Complex Building A

Online:

Zoom Meeting ID: 271 534 5558

Passcode: YMSC

Title:

Simultaneous Inference for Eigensystems and FPC Scores of Functional Data

Abstract:

Functional data analysis has become a pivotal field in statistics, emphasizing data represented by functions rather than scalar values. Although significant progress has been made in estimating fundamental elements such as mean and covariance functions, simultaneous inference for eigensystems and functional principal component (FPC) scores remains challenging.

In this talk, we introduce novel methodologies for the simultaneous inference of eigensystems and the distribution of FPC scores in densely observed functional data, along with the asymptotic properties, especially holding in C[0,1] and for a diverging number of estimators. We validate our approaches through simulations and apply them to electroencephalogram (EEG) data, demonstrating their practical utility in testing hypotheses related to FPCs and the distribution of FPC scores. Finally, we discuss extensions to two-dimensional functional data, functional time series, and a the unified theory bridging sparse and dense functional data.

DATEDecember 15, 2024
SHARE
Related News
    • 0

      Global well posedness of Score-Based Generative model via Sharp Lipschitz estimate

      Abstract:We establish global well-posedness and convergence of the score-based generative models (SGM) under minimal general assumptions of initial data for score estimation. For the smooth case, we start from a Lipschitz bound of the score function with optimal time length. The optimality is validated by an example whose Lipschitz constant of scores is bounded at initial but blows up in finit...

    • 1

      Development of physics-based and data-based molecular simulation methods

      AbstractRecently, molecular simulations have benefited greatly from the development and subsequent application of deep-learning methods. In this talk, we will discuss how machine-learning methods can be combined with enhanced sampling techniques to speed up molecular dynamic simulations. We will then discuss about our recent effort on learning from Alphafold2 to reproduce and improve protein st...