Abstract
The role of recursive residuals in the validation of linear models is well-recognised. Furthermore, their contribution to quality control and change point analysis has become increasingly valued. A new procedure for deriving recursive residuals - assuming data observations are time-ordered - is now presented and its theoretical efficacy established. Relevant results are illustrated across a number of real-life data applications.
Original language | English |
---|---|
Title of host publication | Proceedings of the SMTDA Conference |
Editors | Christos H Skiadas |
Place of Publication | Chania |
Publisher | Stochastic Modeling Techniques and Data Analysis |
Publication status | Published - Jun 2010 |
Event | Stochastic Modeling Techniques and Data Analysis International Conference - Chania, Crete, Greece Duration: 8 Jun 2010 → 11 Jun 2010 |
Conference
Conference | Stochastic Modeling Techniques and Data Analysis International Conference |
---|---|
City | Chania, Crete, Greece |
Period | 8/06/10 → 11/06/10 |
Keywords
- Influence, Leverage, Outliers