Financial Data Science with SAS

Chapter 3: Visual Diagnostics of Financial and Economic Data

Visual Diagnostic of the Salient Properties of Financial Data

Data scientists use visualizations in the exploration phase of the analytics cycle to analyze the statistical properties of the data and to determine how those properties might impact the models under consideration for deployment. This is because most statistical models and machine learning algorithms are trained based on some assumptions about the underlying statistical properties of the data such as normality, random sampling, and homoscedasticity (constant variance). If these assumptions are violated in the data properties, then the data scientist would have to weigh that information in the choice of the models for the ...