Forums  > Trading  > Index trading strategy  
     
Page 5 of 5Goto to page: 1, 2, 3, 4, [5] Prev
Display using:  

NeroTulip


Total Posts: 1010
Joined: May 2004
 
Posted: 2018-04-24 09:17
At what Sharpe does it start making sense to use first loss funding? Not even sure how to think about this rationally... but definitely not at Paulson’s Sharpe LOL. Seems quite desperate.

Has anyone seen the returns of these first loss funds? Must be pretty good if they take half of the gains and none of the losses on a portfolio of high Sharpe strategies.

"Earth: some bacteria and basic life forms, no sign of intelligent life" (Message from a type III civilization probe sent to the solar system circa 2016)

ronin


Total Posts: 267
Joined: May 2006
 
Posted: 2018-04-24 09:43

Step 1: Give lots of leverage to desperate people that nobody else trusts with their money

Step 2: ???

Step 3: Profit


"People say nothing's impossible, but I do nothing every day" --Winnie The Pooh

nikol


Total Posts: 439
Joined: Jun 2005
 
Posted: 2018-04-25 12:10
@Energetic
>> Yes, I've been thinking about this. What bothers me there is that all trades become public so that your strategy could be mined.

This thing always puzzled me. It is like decoding blockchain = you have quasi random string of data, over which you apply a set of coding algorithms (your Technical, Quantitative analysis) with continuous parameters. How it is possible to decode it? Only to regress somehow... We know that many algos (95-99%) work up to transaction cost, where people apply different set/combination of algo-boxes (e.g. in terms of MA, Bollinger, RSI, MACD, SAR, etc etc). I believe that decoding is still difficult...

Perhaps, it is a separate discussion )

Azx


Total Posts: 31
Joined: Sep 2009
 
Posted: 2018-04-25 13:30
@EspressoLover: "Academic finance has done a pretty good job of showing that the sizable majority of day-to-day market variance is driven by changes to the discount rate, not changes to expected cash flows."

Do you have a reference for that? I would expect it to be the complete opposite, that the majority of day-to-day market variance is driven by uncertainty in expected future cash flow. If changes to the discount rate was driving most of the volatility then you could make a lot of money arbing the difference between the variance of daily price movements and variance of realized cash flow. Simply fade all price movements, since any movements on average resulted in a deviation from the expected value.

Maggette


Total Posts: 1019
Joined: Jun 2007
 
Posted: 2018-04-25 14:22
@nikol
The problem is in order to be convincing to your investors, you more often than not give some hints what your strategy is based on and what type of strategy you are using. Mean reverting, trend follwong, long/short, based on blablabla.....

These terms might be rather generic, but combined with the return data it can be quite telling. I think combining the info energetic provided + the return data is probably enogh to reingeneer it.

I would probably use reinforment learning on it :)

Edit: think of the opposite. Often you do find equity curves that don't match the "market neutral" story etc.

Ich kam hierher und sah dich und deine Leute lächeln, und sagte mir: Maggette, scheiss auf den small talk, lass lieber deine Fäuste sprechen...

goldorak


Total Posts: 1028
Joined: Nov 2004
 
Posted: 2018-04-25 15:15
@Nikol: it is perfectly doable and in his strategy's case it is a no brainer. Regressions? I mean investment banks were doing that 10 years ago and usually investment banks are not at the forefront.

If you are not living on the edge you are taking up too much space.

EspressoLover


Total Posts: 296
Joined: Jan 2015
 
Posted: 2018-04-25 21:22
> Do you have a reference for that? I would expect it to be the complete opposite, that the majority of day-to-day market variance is driven by uncertainty in expected future cash flow.

This is basically what Shiller won the Nobel Prize for. Lecture: Using this decomposition and a vector-autoregressive model in difference form, with post World War II stock market returns, Campbell and Ammer found that excess returns innovations have a standard deviation that is two or three times greater than the standard deviation of innovations in future divi- dend growth. Aggregate stock market fluctuations have therefore been domi- nated by fluctuations in predicted future returns, not by news about future divi- dends paid to investors.13 (Figure 2 in those notes also really drives the point home)

> arbing the difference between the variance of daily price movements and variance of realized cash flow.. any movements on average resulted in a deviation from the expected value.

Well, you can exploit it but it's certainly not arbitrage. The duration of equity cash flows is somewhere between 10 and 30 years depending on the year, business cycle, methodology, etc. So you are looking at buying and holding around a decade or longer.

However you can do something kind of like this. An investor can use the CAPE valuation of the market index at the current time to scale his exposure. She can target a consistent Kelly fraction, in which case she'd lever up her equity exposure when CAPE was cheap and deleverage down when CAPE gets expensive. (And possibly if CAPE gets very extremely high like 1999, short equities). This strategy would increase historical returns against plain-old-indexing by a factor of 30-50%.

That being said the percentage of real money sophisticated enough to do this is a drop in the bucket against all the equity exposure in the world. If any phenomenon demonstrates the limits to arbitrage and EMH this would certainly be it.

#Reverse-engineering

I'm going to go out on a limb here, and say that the risk that some malicious party comes in and steals your strategy, then scales it up so much that soaks up all the alpha, is pretty low. You should be much more concerned about the risk that this thing never gets off the ground, because no one really picks up interest. I'd say that scenario is about 100 times more likely than former.

First off, if the returns are anything like historical, 90% of investors are going to more than happy to pay you 2/20 and deal with the day-to-day headaches. Second, this market is so deep and liquid that it would take an extraordinary amount of money to dissipate the alpha. Maybe someone might steal your system and throw a few million here and there, but unless they're quickly willing to scale up to billions on some system they reverse-engineered off a website, it ain't moving prices.

Furthermore, it's not like this is some super-duper secret sauce. People have been publishing on the relation between VX curve and ERP for at least several years now. I do believe that you have specific techniques and features, such that your particular implementation is superior to the published research. But don't think that AQR doesn't have a dozen quants right now working on something pretty damn similar.

Publish the damn code on Github if you need to. If $100 million AUM steals your strategy, and you get $10 million in legit investors, you're still way ahead.

Good questions outrank easy answers. -Paul Samuelson
Previous Thread :: Next Thread 
Page 5 of 5Goto to page: 1, 2, 3, 4, [5] Prev