In this chapter, we presented an intra-day volume forecasting model and its implementation in R using data from the DJIA index. Due to length limitations, we selected the one model from the literature that we believe is the most accurate when used to predict stock volumes. The model uses turnover instead of volume for convenience, and separates a seasonal component (U shape) and a dynamic component, and forecasts these two separately. The dynamic component is forecasted in two different ways, fitting an AR(1) and a SETAR model. Similarly to the original article, we do not declare one to be better than the other, but we visually show the results and find them to be acceptably accurate. The original article convincingly proves the model to be better than a carefully selected benchmark, but we leave it to the reader to examine that, because we only used a short data set for illustration, which is not suitable to obtain robust results.
Mastering R for Quantitative Finance
Mastering R for Quantitative Finance
Overview of this book
Table of Contents (20 chapters)
Mastering R for Quantitative Finance
Credits
About the Authors
About the Reviewers
www.PacktPub.com
Preface
Free Chapter
Time Series Analysis
Factor Models
Forecasting Volume
Big Data – Advanced Analytics
FX Derivatives
Interest Rate Derivatives and Models
Exotic Options
Optimal Hedging
Fundamental Analysis
Technical Analysis, Neural Networks, and Logoptimal Portfolios
Asset and Liability Management
Capital Adequacy
Systemic Risks
Index
Customer Reviews