Timeline for What are the key risks to the quantitative strategy development process?
Current License: CC BY-SA 2.5
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 9, 2017 at 18:17 | comment | added | Swagato Acharjee | data and technological availability is a big one. For example if you do a 10 year backtest, even after correcting for data snooping and overfitting and selection bias many times earlier periods look great, mainly because some of the data you may have used to build your model was not as freely available during that period, as a result the inefficiency was never exploited. Even if the data was freely available some of the algorithms and computational power available to you may today may not have been available at that time to be able to exploit that anomaly in a cost effective way. | |
| Feb 7, 2012 at 8:32 | comment | added | Robert | Nice post Paul, very wise words! I think you should add a link to your article on swap spreads, as a real example showing how you carried out an analysis for a trading idea. I really enjoyed reading it and have learnt a lot, and others will too I am sure. | |
| Feb 3, 2011 at 17:38 | comment | added | Shane | @Zarbouzou Sounds like a good new question to ask. :) | |
| Feb 3, 2011 at 17:24 | comment | added | Zarbouzou | The half life is in fact a big problem to me. Is there any way to estimate it? (even roughly) | |
| Feb 3, 2011 at 17:05 | history | answered | pteetor | CC BY-SA 2.5 |