Abstract:
|
Geographic Information Systems (GIS) and related technologies that are capable of collecting precise positioning information have spawned massive amounts of spatial-temporal data. Spatial "data science" today refers to the use of technology, statistical methods and algorithms to extract knowledge and insights from spatially referenced data. Spatial analysis abound in a variety of disciplines including the natural, environmental, social and medical sciences; economics; forestry; ecology; and public health. This talk will describe different types of spatial data structures, how they motivate different classes of statistical models. We emphasize on the application and many benefits of Bayesian hierarchical models, highlighting some recent developments that harness high performance scientific computing methods for spatial BIG DATA analysis and emphasize on methods that can be implemented on very modest computing architectures (such as a laptop). We will present specific examples of Bayesian hierarchical modeling in Light Detection and Ranging (LiDAR) systems and other remote-sensed technologies; environmental sciences; and accelerometer data in public health applications.
|