How do You De-leverage

[Chris67]

Wondering how others de-lever systems for risk levels they are comfortable with
The obvious answer seems to be to risk less per trade but thi sdoes seem , strangeley (or not) to knock risk metrics down. For example say you design an uber robust system with an MAR of 1.5 that makes 60% per year CAGR and has a 40% max d/d - risking 1% per trade.
So you decide making 30% CAGR with a max d/d of 20% would be acceptable and so risk 0.5% per trade instead but would you believe it teh MAR falls to 1.18 ?
I guess other ways include lowering teh capital by 50% so that you get you 1.5 on half the capital ? I found this method works well - what have others found ?
C

[sluggo]

Have a look at the green curve.

Decreasing the risk-per-trade does indeed decrease the MAR ratio (if you have the good sense to operate on the non-suicidal (left) side of the green dot) as shown on this very website in 2008 (REF)

Attempts to thwart this via continuous rebalancing, encounter the difficult question: how often must I rebalance? Once every three months? Once every hour?

[drm7]

what have others found ?C

I stumbled on a similar discussion on a "quant" trading blog. The author does a lot of stat arb and thinks a lot about optimal leverage vs. performance, etc.

In this post http://epchan.blogspot.com/2011/07/sorr ... or-us.html, he offers a similar solution to yours, with the added twist of using the extra cash as a rebalancing fund. So there is also a "trading the equity curve" factor to consider, because you add funds during drawdowns and withdraw funds after new highs. There may not be a 1:1 translation because he talks in terms of leverage rather than MAR.

2) As an investor, there is an easy way to control leverage and risk: just apply Constant Proportion Portfolio Insurance (a concept also discussed elsewhere on this blog). For example, if the fund manager tells you the fund employs a constant 10x leverage (as dictated by the risk analysis outlined in 1) and you are only comfortable with 5x leverage, just invest half your capital into the fund, and keep the other half as cash in your bank account! Going forward, if the fund loses money, your effective leverage would have decreased to below 5x. Say you invested $1M into the fund, and kept $1M in the bank. And say the fund lost $0.5M. Your total equity is now $1.5M, and the fund manager is supposed to trade a $0.5M*10=$5M portfolio. Your effective leverage is now only 3.33x, well within your tolerance. Now if instead, the fund made money, you can immediately withdraw some of the profits to keep your effective leverage at 5x. So, say the fund made $0.5M. Your equity is now $2.5M, and the fund manager is supposed to trade a $1.5M*10=$15M portfolio. If you don't withdraw, this would increase your effective leverage to 6x. But if you immediately withdraw $0.25M, then the fund manager will trade a $1.25M*10=$12.5M portfolio, giving you an effective leverage of the desired 5x

The comments section of the post has some interesting follow-up discussions. I think that the rebalancing part would add a LOT of work and complexity.[/quote]

[Rush]

[...encounter the difficult question: how often must I rebalance? Once every three months? Once every hour?]

How about re-balancing when needed rather than at a certain subjective intervals (e.g. three months.)?[/quote]

[Aaron01]

[...encounter the difficult question: how often must I rebalance? Once every three months? Once every hour?]

How about re-balancing when needed rather than at a certain subjective intervals (e.g. three months.)?

But then, "when needed" could be described as a subjective criteria as well

[rgd]

I find that measuring total portfolio risk (risk to exit on all open positions), and reducing all positions by a given amount when the total risk number crosses a given threshold, works pretty well.

[Rush]

But then, "when needed" could be described as a subjective criteria as well

It depends.

You can systematically define via rules any logic - as "when needed" - or you can acknowledge that system design is based on certain discretionary assumptions. You can test such assumptions. You can also systematize the process of designing a suit of systems.

It is all relative; it may well be that subjective = objective.

[Bravochico]

You've also got to consider strategy evolution into the mix. When you back test deleveraging strategies it doesn't consider real life strategy evolution. I doubt few traders use all the same strategies over many years. I come up with at least a few new strategies every year and stop using older ones that lose their edge. Jim Simons average strategy life is less than one year. It's an extreme example but nonetheless makes the point. If given the choice of generating good strategies every year versus finding some uber effective deleveraging strategy,I'd take new strategies every time.

When you add this into mix, the simpler the deleveraging process, the better.