[ABE-L] Webinar em High Dimensional Data Analysis

[Ronaldo Dias]

Prezados(as)
Gostaria de convidá-los(as) a assistir mais um seminário em Análise de
Dados em Alta Dimensão. Ocorrerá às 13hs, 02/12/2020. Os links para este
seminário e a gravação do anterior (google meet e youtube) estão abaixo.
Desta vez teremos como palestrante, Alex Rodrigo dos Santos Sousa. IME-USP.
abs.

Webinar 2/12/2020: https://meet.google.com/bzx-pbva-imx

 Model selection criteria for regression models with
splines and the automatic localization of knots

Alex Rodrigo dos S. Sousa

Department of Statistics, University of Sao Paulo, Brazil

Joint work with Florencia Leonardi and Magno Severino from University of
Sao Paulo.


One Model selection criteria for regression models with

splines and the automatic localization of knots of the main issues in
spline regression is the selection of the number and positions of internal
knots. In data analysis, wrong specification of the knots can cause a
high impact
on the spline regression performance and few studies in the literature
provide some systematic method to overcome this difficulty. In this work we
propose a model selection approach to fit a regression model using splines with
a variable number of knots. We introduce a penalized criterion to estimate
the number and the position of the knots where to anchor the splines basis.
The penalty involves the number of segments on domain determined by the
knots and plays the role of avoiding a high number of knots and
consequently to overfit the data. The method is evaluated on simulated data
and applied to covid-19 daily reported cases for short-term prediction.


webinar 25/11/2020: https://youtu.be/jxyn8mqVqCw


-- 
Ronaldo Dias, Ph.D.
Professor
Dept. of Statistics-IMECC, UNICAMP
www.ime.unicamp.br/~dias
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <http://lists.ime.usp.br/pipermail/abe/attachments/20201127/526fbc74/attachment.htm>