macrocephalopod on Twitter: "Or you could make the linear term proportional to volatility and the quadratic term proportional to volatility^2 and get an optimizer which scales the book up and down in response to current volatility conditions (and keeps the benefit of trade buffering and zero net exposure)"

Country	Code	For customers of
United States	40404	(any)
Canada	21212	(any)
United Kingdom	86444	Vodafone, Orange, 3, O2
Brazil	40404	Nextel, TIM
Haiti	40404	Digicel, Voila
Ireland	51210	Vodafone, O2
India	53000	Bharti Airtel, Videocon, Reliance
Indonesia	89887	AXIS, 3, Telkomsel, Indosat, XL Axiata
Italy	4880804	Wind
3424486444	Vodafone
» See SMS short codes for other countries

macrocephalopod‏ @macrocephalopod 13 Nov 2021

"portfolio optimizers always maximize errors" or "optimizers are for dweebs". Wrong! I'll try to explain why. The non-expert view of portfolio optimizers is something like - you have a vector of alphas (expected returns), a covariance matrix, a vector of costs (e.g. spread and

2 replies 0 retweets 37 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

impact), current position x0, and some constraints, and you run a quadratic optimization to find a vector of positions "x" which maximises alpha * x - x' * covariance * x - spread * abs(x - x0) - slippage If you do this naively you will get bad results!

1 reply 1 retweet 35 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

A better approach, the one I use, is to start with a heuristic solution that you know works (i.e. you would make money with it). Then write down an optimization problem that gives you the same result. You're now in a good place because you have a set of parameters for the

2 replies 0 retweets 41 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

optimizer that definitely give a good result, and you can start making changes to the optimizer to try and improve it. One thing optimizers are good at is finding new solutions that are close to existing solutions, but satisfy extra constraints. That means you can change

2 replies 0 retweets 39 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

a parameter, and be sure that the constraints will still be satisfied. That's not generally true if you start modifying a heuristic algorithm - changing one component normally requires changes elsewhere as well.

1 reply 0 retweets 20 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

That's all very vague so let me give an example. You have a vector of signal values "s". Set target positions to be proportional to (s - mean(s)) / sum(abs(s - mean(s))) This has zero net exposure and constant gross book size.

2 replies 0 retweets 19 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

macrocephalopod Retweeted macrocephalopod

You then apply some buffering to it, as described in this earlier thread, to determined what trades you should do. This is the heuristic solution. Note that after the buffering it no long has zero net exposure.https://twitter.com/macrocephalopod/status/1373216917058682881?s=20 …

macrocephalopod added,

macrocephalopod @macrocephalopod

People are saying I am not posting enough finance content. FINE. Once you have decent alphas a lot of the skill in running a systematic strategy is reducing turnover. Alpha is unknown but trading costs are certain. Keep trading to the minimum needed to monetise your alpha.

Show this thread

1 reply 0 retweets 16 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

The quadratic optimization problem that gives this result has a linear term proportional to "s - mean(s)", a diagonal covariance matrix, a risk aversion which is proportional to sum(abs(s - mean(s))) and a simple linear cost term proportional to abs(x - x0)

2 replies 0 retweets 14 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

The simplest enhancement is to add a zero net exposure constraint, or a quadratic penalty on the total net exposure. That gets you all the benefit of trade buffering but also maintains close to zero net exposure - an instant win.

1 reply 0 retweets 12 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

Another direction of research could look at varying the risk aversion parameter while introducing a maximum book size constraint. That will put more weight into the highest conviction positions, while zeroing out the low conviction positions.

1 reply 0 retweets 11 likes

Show this thread

macrocephalopod‏ @macrocephalopod 13 Nov 2021

Or you could make the linear term proportional to volatility and the quadratic term proportional to volatility^2 and get an optimizer which scales the book up and down in response to current volatility conditions (and keeps the benefit of trade buffering and zero net exposure)

6:14 AM - 13 Nov 2021

17 Likes

2 replies 0 retweets 17 likes

1. macrocephalopod‏ @macrocephalopod 13 Nov 2021
  There are ~unlimited ways to tweak the portfolio optimizer, each of which allows you to control one aspect of the portfolio construction and let the optimizer control the others. So have fun, and don't be scared of optimization.
  
  4 replies 2 retweets 33 likes
  
  Show this thread
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. Mohammed‏ @vivaitti 13 Nov 2021
  Replying to @macrocephalopod
  
  Then, the initials guess becomes very important, right? A good sensible initial guess can ensure your Levenberg–Marquardt(or any Gauss Newton equivalent) is fast and actually finds a minimum.
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

Tweets

macrocephalopod