Are quantitative techniques used in distressed investing?
I finished Moyer's Distressed Debt Analysis recently and I'm interested in how the techniques and strategies have changed since it was published. In particular, whether modern AI/ML techniques are used.
I can imagine using NLP for reviewing legal documentation, but I'm not finding much online about other uses.
Any suggested readings or firms to look at?
No
Can you eloborate. Also, are comp and job security usually in line with L/S equity in distressed debt or is it higher. Are hours worse?
Traditional hedge funds use machine learning because it’s useful marketing for increasing AUM. So the purpose of a data scientist is sort of like the purpose of a sexy receptionist or nice office furniture - it makes you seem more credible so you can collect a larger management fee. In the case of distressed debt, we don’t see that effect because people’s first reaction to this idea is “wow, that’s stupid” and not “wow, take my money.” This is the reason why distressed funds generally don’t do quant strategies.
there are plenty of quantitative strategies for trading high yield / stressed credit that I’ve seen or heard of. You won’t see people write about it because it actually is fairly innovative and new. You may not even know who is doing these strategies yet, because some of it is fairly secretive. The keyword here is TRADING and not INVESTING. For the reasons described elsewhere, INVESTING using quantitative methods is not useful given the importance of qualitative information in the decision-making process.
No, if you read Moyer, you should realize a few things that would make it not feasible.
Distressed is a small space. To do proper, quantitative techniques you generally need a large enough sample size for you to be confident in the correlation.
It's very discrete and event driven. Quant is about having a trade that can work >50% of the time and then just repeating that trade over and over so you have a positive EV. In distressed, each situation is so individual and discrete that you cant replicate it. Usually there are only a handful of events per name that actually cause you to make or lose money so the trading in and out doesn't really work.
Distressed is arguably the type of investing that is least prone to quant techniques and automation.
Isn't the junk bond market less prone to quant techniques and automation.
If by junk you are referring to HY, then no....
Both are well protected from automation but distressed is more illiquid and has a smaller set of names so, even though both are highly resistant to automation, distressed is the more protected of those two.
Nam accusantium odio nam quas sed est. Voluptatum amet ut numquam id porro error. Repudiandae ea consequatur exercitationem.
Voluptatem quo explicabo molestiae aspernatur corporis velit quis perferendis. Sunt qui quam modi recusandae voluptates soluta. Repellat debitis blanditiis et expedita sunt.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...