Date:30 September 2020, Wednesday
Location:via ZOOM. Meeting details indicated below.
Time:03.00pm - 04.00pm, Singapore time
In a Data-Generating Experiment (DGE), the observed sample of size n, x=x(θ )(ε Rnxd), has intractable or unavailable c.d.f. Fθ, and θ‘s statistical nature is unknown; θεΘ, d≥ 1. Matching estimates of θ are introduced, from the “best” x-matches with samples X*(θ*), θ*εΘ, obtained from a Sampler or Black-Box with input θ*. Under mild conditions, these estimates are consistent and upper bounds on the rates of convergence in probability (not in distribution) are obtained which, when Θ is subset of Rm, can be (log n/n)1/2; m ≥1. Averages of M Matching Estimates are successful for the mixture 2 normals, Tukey’s (a,b,g,h) and the (a,b,g,k) models. The Empirical Discrimination Index for DGE introduced provides insight on DGE’s parameter discrimination and for comparing DGEs.
This seminar will be held via ZOOM: https://www.zoomus.cn/18430443981/join
Please download ZOOM China from the above link to access the meeting room.
Meeting ID: 18430443981
Password: ICCM
Block S16, Level 7, 6 Science Drive 2,
Faculty Of Science,
National University of Singapore
Singapore 117546
+65 6516 4416
+65 6872 3919
© National University of Singapore. All Rights Reserved.