BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:STA Invited Speaker Series:  A sequential approach to obtain o
 ptimal designs for non-linear models harnessing closed-form solutions for 
 faster convergence (joint work with Suvrojit Ghosh and Koulik Khamaru)
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260520T130714Z
UID:tag:localist.com\,2008:EventInstance_45874637058172
DTSTART:20240322T154000Z
DTEND:20240322T164000Z
DESCRIPTION:Dr. Tirthankar Dasgupta\, Professor and Co-Graduate Director\nD
 epartment of Statistics\, Rutgers University\n\nAbstract:  D-Optimal desig
 ns for estimating parameters of response models are derived by maximizing 
 the determinant of the Fisher information matrix. For non-linear models\, 
 the Fisher information matrix depends on the unknown parameter vector of i
 nterest\, leading to a weird situation that in order to obtain the D-optim
 al design\, one needs to have knowledge of the parameter to be estimated. 
 One solution to this problem is to choose the design points sequentially\,
  optimizing the D-optimality criterion using parameter estimates based on 
 available data\, followed by updating the parameters estimates using maxim
 um likelihood estimation. On the other hand\, there are many non-linear mo
 dels for which closed-form results for D-optimal designs are available\, b
 ut because such solutions involve the parameters to be estimated\, they ca
 n only be used by substituting “guestimates” of parameters. In this pa
 per\, a hybrid sequential strategy called PICS (Plug into closed-form solu
 tion) is proposed that replaces the optimization of the objective function
  at every single step by plugging in the estimates into the available clos
 ed form solutions. Convergence of the sequence of solutions generated by t
 he PICS approach to the true D-optimal solution\, and asymptotic normality
  of the sequence of estimators generated by this approach are established.
  Usefulness of this approach in terms of saving computational time and ach
 ieving greater efficiency of estimation compared to the standard sequentia
 l approach are demonstrated with simulations conducted from two different 
 sets of models motivated by real-life scenarios.
LOCATION:
SUMMARY:STA Invited Speaker Series:  A sequential approach to obtain optima
 l designs for non-linear models harnessing closed-form solutions for faste
 r convergence (joint work with Suvrojit Ghosh and Koulik Khamaru)
URL;VALUE=URI:https://events.miamioh.edu/event/sta_invited_speaker_series
CATEGORIES:Lectures & Presentations
END:VEVENT
END:VCALENDAR