|
Activity Number:
|
458
- Bayesian Methods in Spatial Statistics
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 10, 2022 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
WNAR
|
|
Abstract #323164
|
|
|
Title:
|
Bayesian Data Sketching for Large Spatial Data
|
|
Author(s):
|
Rajarshi Guhaniyogi* and Sudipto Banerjee and Laura Baracaldo
|
|
Companies:
|
Texas A & M University and UCLA and UC Santa Cruz
|
|
Keywords:
|
Bayesian Inference;
B-Spline;
Data Sketching;
Posterior Contraction;
Random Compression Matrix;
Varying Coefficient Model
|
|
Abstract:
|
We introduce Bayesian data sketching for spatial regression models to obviate computational challenges presented by large numbers of spatial locations and to ensure data confidentiality. To address the challenges of analysing such large data, we compress the spatially oriented data by a random linear transformation to achieve dimension reduction and conduct inference on the compressed data. Unlike several popular methods for analysing large spatial data, our approach requires neither the development of new models or algorithms nor any specialised computational hardware while delivering fully model-based Bayesian inference. Well-established methods and algorithms for spatial regression models can be applied to the compressed data. We will discuss posterior contraction rates for estimating the spatially varying coefficients and predicting the outcome at new locations under the randomly compressed data model.
|
Authors who are presenting talks have a * after their name.
