1 / 11

Peter

Looney Tunes Asset Management. Juan. Russell. Kenny. Peter. Assignment Objective. Apply cross-sectional stock selection model to South African stock exchange (100 securities) Identify intricacies of building and running such a model. Model Design. Coded using Excel VBA

romney
Download Presentation

Peter

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Looney Tunes Asset Management Juan Russell Kenny Peter

  2. Assignment Objective • Apply cross-sectional stock selection model to South African stock exchange (100 securities) • Identify intricacies of building and running such a model

  3. Model Design • Coded using Excel VBA • Data and model results stored in MS Access • Excel Solver and correlation tools used • Scalable to any number of securities and regression attributes

  4. Model Flow • Run time series regressions for 100 firms • Store attribute and firm specific coefficients • Retrieve attributes and firm specific coeffs. on a given date for 100 firms • Run cross sectional regression and store intercept and coeffs.

  5. Model Flow • Use coeffs. to forecast next period return • Generate covariance-variance matrix • Optimize portfolio using predicted returns and volatility and store portfolio weights • Automatically repeat process for in-sample test

  6. Database Layout

  7. Problems • Base data incomplete. Extensive rule base included in model to deal with data problems • Low explanatory power of attributes (low t-values) on some securities • Limited data affects regressions • Long processing times

  8. Results • Used book-to-market • Two year in-sample test • Low average R2 = 2%

  9. Proposed Future Improvements • Use of GARCH modeling to forecast volatility • Extend infrastructure to run sorting/screening models • Complete and augment base data with more attributes

  10. Questions?

More Related