Quantitative Stock Selection April 27 2005 The Fuqua

Agenda I. Review of Original Model II. Industry Rescaling III. Extension of Dynamic Factors

I. Review of Original Model The Fuqua School of Business FIN 491 – Quantitative

Factors & Subjective Scores The Fuqua School of Business FIN 491 – Quantitative Stock

Heat Map: EW Returns The Fuqua School of Business FIN 491 – Quantitative Stock

II. Industry Rescaling The Fuqua School of Business FIN 491 – Quantitative Stock Selection

Used North American Industry Classification System to Scale by Industry The Fuqua School of

Rescaling increased return, 21. 4% vs. 19. 3% while decreasing standard deviation, 12. 7%

Overlay portfolio consistently produced positive alpha The Fuqua School of Business FIN 491 –

Takeaways about industry rescaling…. For our model • Intra-industry effects has historically been larger

III. Extension of Dynamic Variables The Fuqua School of Business FIN 491 – Quantitative

Dynamic Model With Incremental Term Structure Varying Weights The Fuqua School of Business FIN

Improved Model Outperforms Over the Period …but Fails in 1999 The Fuqua School of

Overlay portfolio produced positive alpha in fifteen out of seventeen years The Fuqua School

Takeaways about dynamic factors… Examined a number of interaction variables: • Growth and Value

Takeaway about dynamic factors… Time varying “dynamic” weights can be very useful in increasing

IV. Introduction of Fractile Migration The Fuqua School of Business FIN 491 – Quantitative

Preliminary Information • In-sample period: 01 -Dec-87 thru 31 -Dec-98 • Out-of-sample period: 01

Count of Returns The Fuqua School of Business FIN 491 – Quantitative Stock Selection

Average of Returns The Fuqua School of Business FIN 491 – Quantitative Stock Selection

Fractile Stability & Turnover Implications The Fuqua School of Business FIN 491 – Quantitative

$Results & Conclusions • No substantial benefits for long portfolio via analysis of fractile$

V. Possible Areas for Further Analysis The Fuqua School of Business FIN 491 –

Possible Areas for Further Analysis • Incorporate new research into a “supermodel” • Best

Slides: 25

Download presentation

Quantitative Stock Selection April 27, 2005 The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Agenda I. Review of Original Model II. Industry Rescaling III. Extension of Dynamic Factors IV. Introduction of Fractile Migration V. Possible Areas for Further Analysis The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

I. Review of Original Model The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Factors & Subjective Scores The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Heat Map: EW Returns The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

II. Industry Rescaling The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Used North American Industry Classification System to Scale by Industry The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Rescaling increased return, 21. 4% vs. 19. 3% while decreasing standard deviation, 12. 7% vs. 19. 9%!! The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Overlay portfolio consistently produced positive alpha The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Takeaways about industry rescaling…. For our model • Intra-industry effects has historically been larger then inter-industry effect • Gets rid of huge anomaly in our out of sample data for 2003 • Can a middle ground or new weighting scheme be found to capture both inter and intra effects? The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

III. Extension of Dynamic Variables The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Dynamic Model With Incremental Term Structure Varying Weights The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Improved Model Outperforms Over the Period …but Fails in 1999 The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Overlay portfolio produced positive alpha in fifteen out of seventeen years The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Takeaways about dynamic factors… Examined a number of interaction variables: • Growth and Value Regimes • Credit Spreads • Term Structure of Interest Rates Incremental scoring adjustments using yield curve information added the most value to our model. The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Takeaway about dynamic factors… Time varying “dynamic” weights can be very useful in increasing the power of quantitative models: • More information • Interaction effects • Adaptive models However, finding a consistent model can be tricky: • Overfitting • Lots of noise The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

IV. Introduction of Fractile Migration The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Preliminary Information • In-sample period: 01 -Dec-87 thru 31 -Dec-98 • Out-of-sample period: 01 -Jan-99 thru 31 -Dec-04 • Monthly return data • Based upon original model The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Count of Returns The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Average of Returns The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Fractile Stability & Turnover Implications The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

$Results & Conclusions • No substantial benefits for long portfolio via analysis of fractile$

Results & Conclusions • No substantial benefits for long portfolio via analysis of fractile migration … • … Short portfolio, however, tells a different story • Improvements in return, turnover, and transaction costs identified • For further review: • Is new short-portfolio sufficiently diversified? • w/r/t industry? • w/r/t market-cap? • etc. The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

V. Possible Areas for Further Analysis The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance

Possible Areas for Further Analysis • Incorporate new research into a “supermodel” • Best positioning in an overall portfolio • Identification of additional / stronger signals for timevariation of factor weights • Factor interaction with its own turnover The Fuqua School of Business FIN 491 – Quantitative Stock Selection Fuqua Four Bartlett | Foertsch | Kippels | Vance