|
Abstract:
|
More economic data are available for analysis as we move forward in calendar time and as technological progress makes it possible to collect information about more series. Developing statistical methods that can efficiently exploit the large volume of information available is therefore important. In this session, we will discuss statistical methods for analyzing large panels of economic data. While principal components is often used to reduce the dimension of economic time series, methods such as regularization (e.g., LASSO) and adaptive learning (e.g., boosting) are more popular in machine learning and gene classification. We will discuss the strengths and weaknesses of these methods with emphasis on their role in economic forecasting and financial analysis.
|
