A.I.M. Users Bulletin Board

[Tim Reese]

Hi Conrad & group - you wrote

I have copied the article you referred to but have not yet digested it fully. Er, the article I was thinking of is gone, but it's successor (I think) is on
http://www.gmo.com/mygmoasps/Research.asp?Marketcommentary=True&AssetID=009AsstAll

GMO is an interesting firm, seeming to have at least one foot in the quantitative investment camp. They quantitatively manage one of my mutual funds, VUVLX.

I would be interested in the results and any spread sheets - I can be reached at

[e-mail]

Obvious how to change the address ...

When I commented on the backtester, I wanted to convey that I was mildly dissappointed, because I did not see wild swings in the results for that single data set when I changed the obvious knobs, such as cash fraction, SAFE, PC add fraction, etc. A very limited test using a very bland price history.

... it should be done formally by experienced testers. Shoot, I haven't let ignorance or inexperience deter me yet <g>, so I'll press on.

In my opinion the objective is not so much testing the effectivities of different AIM Structures but the idea of automatic optimisation(various ideas can be incorporated into this goal. As I see it we should define exactly what it is we want to optimise and how we are going to do it, if we are going to do it. I presume this means dynamic as well as automatic optimization - all fine within the AIM mandate as long as we set the rules in advance. This could also include rules for allowing or calling for judgement and intervention, ie when to bail, when to reset, etc.

More dinking around with toy programs this afternoon

best regards Tim