Posts with #FunctionalData

#FunctionalData

Public posts on X that use #FunctionalData.

Related hashtags

  1. Omar Rivasplata✓@OmarRivasplata
    I've seen #FunctionalData papers they talk about "random functions" sampled from a "process" but without precise definitions for these objects. Anyone familiar with the literature can clarify the meaning or point to references giving precise definitions?
    Download
    112401
  2. Dolores Romero Morales @doloresromerom.bsky.social@DoloresRomeroM
    Check in @ResearchGate our new preprint on #optimaltrees and #functionaldata with @needs_project PhD student @crismolerorio from @imus_us, @RafaelBlanquero and @emiliocarrizosa Happy 🔍🔍🔍! #orms
    Cristina Molero del Río @crismolerorio
    📌 Check out our new preprint! 📄 On optimal regression trees to detect critical intervals for multivariate functional data 👥 @RafaelBlanquero @emiliocarrizosa @DoloresRomeroM 🔗 @imus_us @unisevilla @CBS_ECON @needs_project #riseneeds #ORMS
    016
  3. Dominik Liebl@domliebl
    Hey #statstwitter ➡️ Here's an interesting webinar on: "Recent Advances in Functional Data Analysis" Organized by @YoungStatS2 Further Information (incl Zoom-Link) here: 1/n
    168
  4. Dominik Liebl@domliebl
    @domliebl
    Papers: Alessia Caponera: "Asymptotics for spherical functional autoregressions" doi: 10.1214/20-AOS1959 Fabio Centofanti:"Functional regression control chart" doi: 10.1080/00401706.2020.1753581 Alexander Volkmann:"Multivariate Functional Additive Mixed Models" arXiv:2103.06606
    111
  5. Dominik Liebl@domliebl
    @domliebl
    In case you haven't (yet) heard about "Functional Data Analysis" (FDA) #functionaldata ... ... here's a link to an introductory crash course I once taught ✌️: 3/n
    114
  6. Dominik Liebl@domliebl
    In principle, I like this message of @johnharry76: "Smoothing your curve data, can harm your subsequent analysis." Very true! But, there are some issues. Here's constructive feedback. Let's go. 1/n
    John Harry, PhD @johnharry76
    More from our @TTU_Biomech team's @NSCA 2020 posters: Stop🛑all the fancy low pass filter processes when studying GRF curves from the CMJ. They compromise your results! But.... if you're like me and can't not filter😬, a 50 Hz cutoff is the way to go!
    113
  7. John Harry, PhD✓@johnharry76
    More from our @TTU_Biomech team's @NSCA 2020 posters: Stop🛑all the fancy low pass filter processes when studying GRF curves from the CMJ. They compromise your results! But.... if you're like me and can't not filter😬, a 50 Hz cutoff is the way to go!
    Download
    3744
  8. Dominik Liebl@domliebl
    @domliebl
    In order to further reduce the bias-problem in your data, you need to take into account the varying curvature of your curve data. That is, use a varying ("local") smoothing parameter, instead of a global smoothing parameter (global=the whole domain [0,1.4]). 5/n
    101
  9. Dominik Liebl@domliebl
    @domliebl
    Sorry for the lengthy writing. But I do statistical research on curve data (#functionaldata) and I have a soft spot for #sportscience. So, I could not do otherwise. ✌️ Hope this helps. 6/END
    102
  10. Show more posts