Forums  > General  > Are we in a quant meltdown akin to August 2007?  
     
Page 1 of 1
Display using:  

EspressoLover


Total Posts: 337
Joined: Jan 2015
 
Posted: 2018-10-31 17:54
Some of the numbers coming out of equity hedge funds are looking really ugly this month. Obviously we've suddenly fallen into a global equities correction. I'd expect the long/short funds to get clobbered as they're mostly going to be long biased in a late bull market. But relative to their historical range, it looks as if the equity market neutral funds are doing even worse. So it's not just a general issue of losses on beta.

I know earnings responses have been really wonky, and that's probably fucking up a lot of interday quant strategies. Also sector rotation volatilty has spiked faster than index vol. At least from the 10,000 foot perspective, things are kind of looking like August 2007, or at least August 2011. That is normally placid quant factors going through violent moves because of synchronized de-risking by the major multistrats.

But this is totally just supposition, and I currently have very little transparency into the day-to-day of this space. Was hoping other phorumers with insight could add some color, or at least join in the unfounded speculation.

Good questions outrank easy answers. -Paul Samuelson

Strange


Total Posts: 1450
Joined: Jun 2004
 
Posted: 2018-11-01 02:26
Indeed, I saw the MtD numbers from a large PB today and wanted to hug myself.

On the other hand, this feels very anti-vol, i.e. people are not reaching for protection.

I don't interest myself in 'why?'. I think more often in terms of 'when?'...sometimes 'where?'. And always how much?'

tradeking


Total Posts: 18
Joined: May 2016
 
Posted: 2018-11-01 03:01
Not for all quant strats. Christmas came early for HFT traders.

Craig


Total Posts: 41
Joined: Oct 2008
 
Posted: 2018-11-01 06:38
Anybody providing liquidity is doing alright, I don't think anybody knows what will happen, but I'd be happy for the current volatility to continue.

ronin


Total Posts: 361
Joined: May 2006
 
Posted: 2018-11-01 10:57
It will be interesting to see the overall numbers.

There are a couple of reasons why market neutral is suffering. Vol is up, correlations are off. And, beta. Market neutral doesn't imply beta neutral, what ever people may think.

"There is a SIX am?" -- Arthur

Maggette


Total Posts: 1062
Joined: Jun 2007
 
Posted: 2018-11-01 13:02
Ok you lost me here
"Market neutral doesn't imply beta neutral, what ever people may think."

Could you please expand on this a little?
Thanks in advance

Edit: ok never mind. I think I got it.

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

ronin


Total Posts: 361
Joined: May 2006
 
Posted: 2018-11-01 14:05
There are high beta stocks and low beta stocks.

E.g. most of the SPX performance in the last 2 years famously came from 5 tech stocks. So of course they are high beta, and of course you want them in your longs. But they especially got hammered in October.

And it's a general thing. Because of how index rebalancing works, there is lots of overlap between performers and high beta stocks. So good market neutral strategies tend to have positive beta bias, unless they specifically target zero or negative beta. But targeting zero or negative beta tends to kill the performance.


"There is a SIX am?" -- Arthur

anonq


Total Posts: 13
Joined: Aug 2018
 
Posted: 2018-11-01 15:42
Luckily not looking like August 07, was on a stat arb desk at an IB that dropped well over 100M during that, crazy times.

October was ugly, we were down going into yesterday but saw a huge bounce as did others and ended up positive for the month and having another good day today.

I think it’s just a crowded space where the unsophisticated and sloppy models (taking on factor bets unbeknownst to them) are having an especially hard time, well modeled shorter term stuff and especially hft still doing great. We run beta and dollar neutral but some explicit factor bets that were good for years haven’t been working for us lately. I guess the easy money can’t stay easy forever, so it goes

Strange


Total Posts: 1450
Joined: Jun 2004
 
Posted: 2018-11-02 03:50
@ronin "Vol is up, correlations are off."

If anything, vol is not up enough considering the magnitude of the moves. I'd say there were eager sellers of vol throughout this turmoil.

I don't interest myself in 'why?'. I think more often in terms of 'when?'...sometimes 'where?'. And always how much?'

Maggette


Total Posts: 1062
Joined: Jun 2007
 
Posted: 2018-11-02 10:54
Thx ronin!

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

ronin


Total Posts: 361
Joined: May 2006
 
Posted: 2018-11-02 11:09
> If anything, vol is not up enough considering the magnitude of the moves. I'd say there were eager sellers of vol throughout this turmoil.

@strange, you sure? this is vix vs 1m and 3m realised. Vix overshoots to roughly 1m realised plus 10, then relaxes. Looks pretty usual.




But yes, it's far from the 2008 vix-hitting-70 thing. Presumably that is still to come.


"There is a SIX am?" -- Arthur

Strange


Total Posts: 1450
Joined: Jun 2004
 
Posted: 2018-11-02 13:40
Well, it obviously depends on the metric. The skew softened more than it should given the level of vol, also most of the days vol was underperforming the skew. The term structure was pretty well behaved most of the days. Also, several times during the selling days you could see people smacking the front to buy the back. Vol of vol got bid up early on only to get softer later on.

I don't interest myself in 'why?'. I think more often in terms of 'when?'...sometimes 'where?'. And always how much?'

Igor


Total Posts: 11
Joined: Nov 2014
 
Posted: 2018-11-03 03:25
My 2 cents from a quant equity market neutral perspective.

In global DM, October saw a large rotation out of growth, (generalized) momentum and into size, value, quality. The former group significantly outperformed the latter in recent years. For example, global price momentum outperformed value by ~15% YTD prior to October. Currently, it's ~8% YTD.

The rotation was rapid and exacerbated by several factors.

1) Equity risk factors are very crowded in today's age. Inter-region correlations of most single equity factors has risen in the past 5 years. An abundance of global assets chasing the same premia.

2) Macro regime shifting. Reversal from the abnormally high liquidity, high growth, low vol, low inflation state of recent years.

To EL's point on synchronized derisking, I'd call it a synchronized rotation out of risk-on and into risk-off factors.

"..tempus casumque in omnibus"

ronin


Total Posts: 361
Joined: May 2006
 
Posted: 2018-11-03 22:06
@strange, either that or your slv parameters need recalibrating....

@igor,

Fair enough. But that rotation also happens naturally when the air is let out of everything - there is more air in growth than in value, so they rotate. I don't think one month is enough to come to any sweeping conclusions. I wouldn't necessarily go long value now.

"There is a SIX am?" -- Arthur

Igor


Total Posts: 11
Joined: Nov 2014
 
Posted: 2018-11-04 02:54
@ronin agreed. Perhaps rotation was a strong word, but I didn't mean to imply the start of a new trend. Rather, a reversion of extreme factor divergence.

"..tempus casumque in omnibus"

ronin


Total Posts: 361
Joined: May 2006
 
Posted: 2018-11-12 11:42
A propos nothing:

Quants Are Facing a Crisis of Confidence

The note was in early September though, not October.

"There is a SIX am?" -- Arthur

Strange


Total Posts: 1450
Joined: Jun 2004
 
Posted: 2018-11-12 14:46
If October results are what people say they are, "crisis of confidence" is an understatement for the current state of affairs.

I don't interest myself in 'why?'. I think more often in terms of 'when?'...sometimes 'where?'. And always how much?'

FDAXHunter
Founding Member

Total Posts: 8367
Joined: Mar 2004
 
Posted: 2018-11-12 15:52
You guys are probably aware, but October/November was one of the ugliest periods on record for Trend Followers in the CTA universe (which are all systematic), with the SG CTA Index delivering -5% over 40 calendar days.

The Figs Protocol.

ronin


Total Posts: 361
Joined: May 2006
 
Posted: 2018-11-12 18:43
Well, 5% monthly volatility isn't that bad. It's only 17.3% annualised. Over 40 days, its 16.5% annualised. Higher than the industry average, but no big deal.

The only problem is if you've had a bad year, bad two years, bad five years and bad ten years - then a -5% month is easily the final nail in your coffin.

But yes, it's a bit frustrating when you are fundraising. We are calling up allocators and telling them here is our fund, the best performing equity market neutral product on the market, and they are reacting like we are saying here is syphilis, the best venereal disease in the world...

"There is a SIX am?" -- Arthur

Strange


Total Posts: 1450
Joined: Jun 2004
 
Posted: 2018-11-13 06:14
Can we expand this to call it a "hedge fund meltdown"? Saw the final MtD numbers for October and every style out there got killed. Not sure who is making money out there (is it only us, "Long Bias Vol RV"?).

I don't interest myself in 'why?'. I think more often in terms of 'when?'...sometimes 'where?'. And always how much?'

tabris


Total Posts: 1256
Joined: Feb 2005
 
Posted: 2018-11-13 09:32
@Strange: what numbers did you see? From the numbers I am seeing seems like they are on average down 1-2%... so not really as bad as the peers being discussed here considering their own respective monthly vol

Dilbert: Why does it seem as though I am the only honest guy on earth? Dogbert: Your type tends not to reproduce.

HitmanH


Total Posts: 463
Joined: Apr 2005
 
Posted: 2018-11-13 10:39
People had a huge bump on the 31st - which made a lot of numbers come out a lot nicer...

Strange


Total Posts: 1450
Joined: Jun 2004
 
Posted: 2018-11-17 18:29
One thing I overheard yesterday was that people seem to be losing money regardless of direction, which is very indicative of system-wide deleveraging.

I don't interest myself in 'why?'. I think more often in terms of 'when?'...sometimes 'where?'. And always how much?'
Previous Thread :: Next Thread 
Page 1 of 1