Ole baby: Phew! Am I glad you like me! I have seen what you do to people who you apparently don't!! But, never in my wildest dreams would I have believed you have a finger and toe fetish! And, that was the reason why.
I don't do regression on the XAU. The only US indices I analyse are the Dow and Trannies. All the same, you do mention a very important point and that is consistency in the data. Clearly, if a company or an index has changed significantly, for any reason, then the assumption of continuity in the analysis is no longer valid. In the circumstance, one would then require a long time period of new data for the changes to exert a significant influence on the outcome. As it is, for the models to have any "predictive" accuracy, the data has to be at least 14 years old. So, one can become quite ancient playing this game. And then, the game changes!
For the reason you mentioned, the data of the SAf gold shares are now messed up. As you know, the important mines have all changed and merged and what-not and therefore the new data are no longer continuous with what went on before. So it's very difficult to keep consistency although, it seems, my "pooled" indices are still working.
Apropos US golds, I do analyse SAf gold share prices (as if they were quoted) in $US and also (and you will find this interesting) as if they were quoted in KrugerRands --- which, I'm sure you'll agree, is also a currency. I thought that was very cute --- evaluating gold shares in units of gold.
I don't know what you like so I'll send you a fingerprint! |
