AIM UK

[ls7550]

Hi Wim

to compare Ladder with AIM and/or LD-AIM, did you have make some backtests?

Over the years I've stripped down and tested what seems like every possible component and combination of AIM.

Don Carlson's EZM was the precursor to Ladder, coupled with some D'Alembert betting sequence style and synthetic virtual Call/Put Options that I was evaluating at the time. Don did do some extensive EZM backtesting and found Level Cost (the style Ladder is based upon) to be the better choice.

Ladder is similar in many respects to LD-AIM in that you can go 100% cash (unlike conventional AIM). LD-AIM typically scales trade sizes by using virtual stocks to increase the minimum trade size (as a percentage of stock value). Ladder scales the trade size by increasing cash - on the basis that two Ladders individually track stocks that don't beat to the same drumbeat.

With both LD and conventional AIM I was continually evaluating at what price level cash would likely be exhausted and I had concerns about the dynamics of maintaining accounts (adding and reducing account sizes). I also wanted to be able to tune trades to my own preferred levels (for instance for market wide index funds I like to trade after a 5% price move in the same direction as the previous trade or 10% price moves in the opposing trade direction).

Generally Ladder fits in better with my overall money management and trading style. Typically however I have lower lows at which cash burn is supported than does conventional AIM and accordingly the (non scaled) trade sizes and hence volatility capture gains tend to be smaller. Equally though Ladder can be scaled to larger trade sizes by narrowing the top/bottom price range (for instance you could perhaps tune the top/bottom range to Bollinger Bands).

With Ladder you have three primary variables that you can adjust, the top, bottom and total amount of funds allocated. I feel that Ladder provides a better instant overview when adjusting any/each of these that does conventional AIM.

In view of such variables it would make backtesting somewhat meaningless. I could for instance backtest using fixed variables but then that would just present one possible result for one particular type or price motion waveform.

So in short no I haven't performed detailed backtests to compare Ladder with LD/AIM, neither do I intend to. I'm 'testing' in the forward direction using real funds ;>)

Regards. Clive.