A Random Intercept Model using Constrained Regression Splines with Application to Tree Height Estimation

Event Date: 

Wednesday, June 6, 2018 - 3:30pm

Event Date Details: 

Refreshments at 3:15 pm

Event Location: 

  • Sobel Seminar Room; South Hall 5607F
  • Department Seminar

Title:

A Random Intercept Model using Constrained Regression Splines with Application to Tree Height Estimation

Speaker:

Xiyue Liao (UCSB)

Abstract:

We propose a random intercept model, in which the fixed effect is estimated by regression splines with shape constraints. The random-intercept component is estimated by the best linear unbiased predictor (BLUP). An optimal convergence rate is established for the fixed-effect component.  Inference methods are discussed for both components. Simulation results about the proposed methods and a tree-height data example are included.