Some High-Diemnesional Techniques for Analyzing Spatial and Other Complex Economic Data

• Activity Number: 274 - Macroeconomic Forecasting and Policy in Data Rich Digital Age Environments
• Type: Invited
• Date/Time: Tuesday, July 30, 2019 : 8:30 AM to 10:20 AM
• Sponsor: Business and Economic Statistics Section
• Author(s): Taps Maiti*
• Companies: Michigan State University
• Keywords: high-diemensional Regression; Post-model selection; Lasso; Group Lasso; Variable Selection; Econometrics

Abstract

Spatial regression models are important tools in economics and social science study. However, the methods are not well developed for small sample big data situation. In this talk we will discuss two topics in this context. First, we talk about high-dimensional spatial econometric models. We argue that these methods generally good for estimation. However, they may not be appropriate for classification or when the dependence is not explicit. Then, we propose high-dimensional machine learning techniques which can be applicable in wide range of applications.