Weekend Wars: Qualitative vs. Quantitative
About two years ago, I had a fascinating market related conversation with the son of a high ranking Chinese Communist Party member who had recently finished a top 5 MBA program stateside.
He was celebrating all over town and the entourage grew by the bottle. By the time I bumped into him, the next destination was already determined...
Enter at your own Risk
Sadly, due to the potent mix of Grey Goose and that lotion/perfume one can only inhale within the bulge bracket of gentlemen's clubs, I forgot not only the kid's name, but my Crackberry and scarf in one of the booths...
Oh well, so much for most favored Capitalist Devil status... I had my Shanghai trip all planned out...
Upon our introduction, he was very cautious and tight lipped. By the time we reached our secret location, the Goose had done it's thing. He was revealing his top secret S&P 500 model and by drink #5 and dance #6, some girl named Heaven or Heather or Heath Bar had herself the next six months worth of rent taken care of.
But here was the bone of contention:
I contended that his model was worthless due to its reliance on past results. I argued that volatility parameters had changed forever and that no amount of statistical data from the past could predict the future. Especially in late 2008.
While spitting all over everyone and everything in sight, my new comrade extolled the virtues and wonders of his infallible quantitative approach. Apparently, he had crushed all competition in every math contest he'd entered since childhood and firmly believed his model could predict Vikram Pandit's next bowel movement.
When I asked him how he modeled for fear...panic...human emotion...the whole club went still for a split second.
I felt like I was about to be garbage-bagged and stored in a Haier.
I'm not sure I understand your question...my data includes 25 years worth of S&P pricing, fluctuations, shifts, the volatility has been priced in to all forecasts and my predictions constantly beat the market...I mean seriously...when have you ever made money off of panic or human emotion?
He added a smirk to his self-congratulatory chuckle...I decided not to pursue it. The rest of the night was fun... the little I remember.
What got to me post-mortem was that this guy will definitely be a player in global economics someday...may already be... I began to wonder how many future finance world decision makers had such tunnel vision.
What do you guys expect to see as the investing approaches and strategies of the coming decade(s)?
I know it's complicated...and no I don't want a stock tip...
I just can help to wonder...
What awaits a system...organized around numbers, but driven by emotion?
I'd first like to say:
FIRST!
The distinction between models and reality should be obvious to anyone who's done hard science. That's why the best quantitative minds spend so much time worrying about when/how/why the models break down, and making adjustments. If you find a strategy that's consistently profitable, then you hit it as hard as you can as long as you can. Hopefully u're ready to move on when/ if it breaks.
There is no quantitative vs. qualitative argument. Super quants and super fundamentalists can both make huge amounts of money really quickly (though I envy the consistency of a quant with a good niche).
Anyone that ignores the human aspects of the market is myopic. If it turn out this kid is an unparalleled quantitative mind ala Jim Simons then I guess it wont matter, but all else equal he'd be better off with a little intellectual flexibility.
I like this, because both can be profitable.
And i believe in mean-reversion to its fullest extend.
Get out of the way of any 3sigma move and watch the trend come back to whatever it was prior.
Did you come up with that last sentence or is it a quote?
yes i did
Nice post. This is actually something I think about a lot. As an econ major, I became increasingly convinced that a quantitative approach to understanding the world is the way to go. So much so, in fact, that I headed to grad school to learn a shit-ton of arcane, esoteric models of human behavior. A funny thing happened in grad school actually; as most of my peers were becoming increasingly convinced of this quantitative approach, I was actually going in the opposite direction. My experience there actually convinced me of the huge limitations of a quantitative approach to the social sciences. I'm not saying you don't want any quantitative models, but rather that people should realize that there are huge limitations to this appraoch, and there's probably much deeper insights in the work of primarily literary economists (Adam Smith, F.A. Hayek, etc). It kind of reminds me of Gresham's Law, but instead of "bad money driving out good," we have a lot of bad economics driving out a lot of good stuff; as economics has became more and more quantitative, economists have seem to forgotten a lot of the insights of more qualitative folks. Well, let's not throw the baby out with the bath water. Lastly, one of the problems with this quantitative approach, is that it makes people think it's more scientific than it actually is. Hayek referred to this as "scientism." Well let me tell you, just because modern economic models looks a lot like physics, doesn't mean they're anywhere near as accurate.
3 years of studying econ and this is word for word what i want to write on the cover of my final exams.
Couldn't explain my opinion of econ any better. I entered college loving econ but by the time I graduated I hated it, as all my classes were purely mathematical and didn't encompass the qualitative aspects of the subject. I felt like by taking such a quantitative approach certain aspects of human nature were totally overlooked.
The word "Model" tends to be over used. I wouldn't always assume that the words model and quantitative methods are synonymous.
Midas, I should give you a -1 paying more attention to model theory than to "model theory" and by the latter I mean the type of model that accompanies your bottles... of goose... and in this case the model happens to be topless... I know they don't technically have souls but you at least could have remembered her name...
All in all it is an interesting topic though... personally, I think the guy must have missed the real value in his top 5 education because simply looking at 1929, 1987, Tech crash, most recent housing crash, it is easy to tell that the human element absolutely causes fortunes to be won and lost. Sure, I would rather have a perfect model to account for everything, but if that were possible LTCM would still be around and conversely, if the human element weren't so vital to our market, then Soros wouldn't have been able to single-handedly devalue the pound.
One of the biggest shortcomings of a pure-quant approach is that it does not account for the emotion of the crowd.
that wouldn't be pimpin... pimpin
This is a very interesting topic that I have been interested in some time now and have read a bit on.
What is interesting to me is to see these model's and how they quantify as you have mentioned non quantitative inputs (essentially all of human emotion).
To provide some insight on your original question, I think we are going to see more of a balance in the 'new' economic system.
I would say there is a correlation with consumer spending and market growth/contraction (pretty obvious). To be honest marketing and the frivolous way of thinking which is a by product of mass media and relates directly to the logic of spend now, well pay later has not worked out as we have seen (ie negative savings rates, over leveraging etc).
I think it has had a big impact on the AVERAGE consumers spending habits (meaning, excluding everyone on hear) because I don't think we realize how ignorant and easily dupet people are. When I hear a girl in college say "Come one lets go to Miami for spring break" and her friends response is "I can't I am broke" and the first girls replys back with "O come onnnnn just get another credit card". Acting like its free money... that when things start to take turn for the worse, and the even more shocking part is they learned these spending habits from someone and they will pass them on to someone down the road as well. I bit worry sum if you ask me.
However one positive thing we can take away from this is that people in North America are starting to open there eyes now and realize that spending out of economic troubles is not the solution anymore. What I think will happen in the US is that human emotion and rational will start to weigh more(than it has) on purchasing and spending habits, because they will start to have to make concession as to what they can do and how often they can do it.
This is just my small insight into a very broad topic that I find very interesting.
I always felt like Economic History would make for a far better curriculum in the real world preparatory sense.
Unfortunately, I haven't seen it offered anywhere other than as a PhD. This is just one of the many examples (IMHO) of why finance professionals are being prepared to fail in the real world, by an inept academic orthodoxy.
You should do an entire blog post that expands on this point. Seriously, I think that would be a pretty epic topic/discussion.
You know I've thought about this a few times actually. Feel free to chime in with some of your experiences on the subject if you'd like.
Once we are smart enough to analyse human behaviour in small increments and come up with some complete graph-theoretic representation oftheir overall behaviour, and we are able to connect this to our neuro-economic models, or neural modals in general, and model the inter-individual behaviour and group behaviour with some nice other quantitative model,
it will really no more be about predicting the behaviour, but predicting the likelihood of the possibility of inducing that behaviour.,
MUAHAHAHAHAHAHAHAHAH
... But seriously, qualitative analysis? I mean, if you are seeing numbers in different colours, you might blast all you got at the share-collections that creates the Mona Lisa, but maybe ... I don't know... qualitative analysis? Reminds me of good ol' policework and that gut feelin' they all talk about. ....
Won't say it's an exhaustive reference, but reading "The Worldy Economists" is a nice starting point for understanding where economics arrived and then acting as a springboard forward. I have no real interest in esoteric economic niche, but I as a non-econ major in undergrad, found the book instructive.
Nice post. Damodaran recently expressed a similar opinion (http://aswathdamodaran.blogspot.com/).
I don't believe complex models are the answer, but I suspect we will see many more of them. My humble opinion is that with the increase in number and complexity of models, we may actually see greater market inefficiency.
So on average, over a loooong time period, what do you guys think is going to make more money?
qualitative or quantitative?
INFORMATION will always make the most money. The qual vs. quant argument is moot in reality b/c it's impossible to track the diverging success/failure rates in real time...and as we know hindsight is always 20/20.
The KEY POINT in my OP that seems to have gone over everyone's heads (or perhaps beneath their feet?) is that this kid is PLUGGED. Therefore, there's no difference for him and anyone sitting atop the oligarchy.
He'll always have access to the sort of info that moves markets and makes money. Therefore, which "God" he prays to is totally inconsequential for him. A guy from this sort of background 40 years ago would have looked at you just as strangely if you said computers were the future.
As for us mortals, I think it's going to be very skewed in an odd way. Namely, nobody can deny that everything is moving in the quant direction, HOWEVER, unless you are a skilled quant/programmer with access to the sort of money that allows you serious leverage (i.e. top prop/HF employee) it is HIGHLY UNLIKELY that you will be able to take advantage of the quant modeling to a full degree. As a result, I think a new form of qualitative analysis will appear which is largely based on a textbook quant approach combined with the individual's qual skills. What this hydra will look like, I can only guess...
Just some thoughts that may add to the conversation...
Lets be honest, quantitative analysis is very useful and accurate to a degree. But qualitative analysis has the certain 'je ne sais qui' and in my opinion requires a really good grasp on the over all market to be able to make a good prediction on it. Call it a gut feeling if you will, but at the end of the day if you have a proven track record of being able to input qualitative criteria into your way of thinking/analysis/model your doing something right.
Soley doing qualitative analysis will get you know were, on the other hand the market has become so technical that quanitiative analyis has in the recent past 2 decades provided the amazing results, but now... and for what I suspect will/can happen is that its going to be more of a mixture of the two.
Excepturi rerum repellendus doloribus nam porro voluptas. Libero autem voluptas qui doloremque est.
Impedit non occaecati tempore repudiandae omnis laboriosam. Possimus et rem consequuntur officiis.
Quos suscipit rerum nisi qui iusto. Ut voluptatem quaerat cumque suscipit veniam beatae consectetur. Omnis voluptatum neque totam repellendus voluptas exercitationem. Earum eum adipisci quam dolores. Placeat ut ab quia voluptate est et sequi. Hic aliquid voluptatem labore.
Ducimus debitis est deserunt sunt. Repellendus nisi debitis ut aut.
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...
Porro ea laborum dolor unde architecto veritatis. Alias consectetur suscipit facere et aut. Qui ipsam quaerat dolore totam exercitationem laudantium. Provident magni dignissimos doloremque dolorem autem libero excepturi. Saepe alias dignissimos est.
Dignissimos quae atque aperiam quaerat. Dicta enim sint blanditiis facilis. Voluptatem libero magnam nihil sed et delectus. Repellendus quia accusantium ut ex est voluptate explicabo.
Pariatur harum dolor rem tenetur ratione distinctio voluptas ratione. Facere non tenetur repudiandae ad qui et occaecati. Porro aut dicta ipsum ipsam quasi numquam similique. Et id ut similique aperiam et ipsam deserunt. Iste quia sint corrupti qui earum omnis.