"Trading as a Business" quant interview, highly recommended

[pedro01]

Quote from TraderZones:

Before making wild assumptions about what caused the crash of 1987, you need to have a basic grasp of economics.

As I recall, interest rates were doing wild things that year prior to the crash, that precipitated it. I don't have the 87 bond or index charts up, but I believe that was the year that happened. It was not a surprise at all, and had little to do with quants or mathematicians.

I made $5K that day, by loading up on mutual funds at EOD prices. I figured it was a serious overreaction, and I cashed in.

But your analysis is seriously weak. As I said before, unsupported opinions are fertilizer.

I don't disagree with you in terms of the trigger.

Portfolio insurance dictated that you sell your positions as the prices fall - of course it's more complex than that. In 1987, the market fell 23% in one day. Of course, there had to be a trigger for people to initiate their portfolio insurance trades.

The problem obviously is that people leant on the portfolio insurance as it effectively told them that they had little risk because they could simply issue the relevant sell orders as prices fell and they would be well covered.

What actually happened of course is that everyone sold at the same time, prices fell and portfolio insurance dictated that they sell again and again and again. There you have excessive risk combined with a model that told everyone to sell at the same time with the obvious consequences.

Similarly of late - there could have been a lot of Puerto-Rican gardeners with $14K salaries ad $1.5 Million mortgages default and the world would barely have noticed. At the heart of this crisis was the Gaussian Copula formula.

Gaussian Copula was a clever bit of math which claimed to model default correlation without looking at historical data but instead used current market data from credit default swaps. So - instead of using historical data (also inherently flawed), the correlated risk of mortgage defaults was calculated based on the price of another instrument. In terms of correlated risk - I'm going to borrow from an article published by a friend :

To understand the mathematics of correlation better, Salmon said that we must consider something simple, like a kid in an elementary school: Let's call her Alice, he said.
“The probability that her parents will get divorced this year is about 5 percent, the risk of her getting head lice is about 5 percent, the chance of her seeing a teacher slip on a banana peel is about 5 percent, and the likelihood of her winning the class spelling bee is about 5 percent.
If investors were trading securities based on the chances of those things happening only to Alice, they would all trade at more or less the same price.
But something important happens when we start looking at two kids rather than one—not just Alice but also the girl she sits next to, Britney.
If Britney's parents get divorced, what are the chances that Alice's parents will get divorced, too? Still about 5 percent: The correlation there is close to zero.
But if Britney gets head lice, the chance that Alice will get head lice is much higher, about 50 percent—which means the correlation is probably up in the 0.5 range. If Britney sees a teacher slip on a banana peel, what is the chance that Alice will see it, too? Very high indeed, since they sit next to each other: It could be as much as 95 percent, which means the correlation is close to 1.
And if Britney wins the class spelling bee, the chance of Alice winning it is zero, which means the correlation is negative: -1.
If investors were trading securities based on the chances of these things happening to both Alice and Britney, the prices would be all over the place, because the correlations vary so much.
But it's a very inexact science. Just measuring those initial 5 percent probabilities involves collecting lots of disparate data points and subjecting them to all manner of statistical and error analysis.
Trying to assess the conditional probabilities—the chance that Alice will get head lice if Britney gets head lice—is an order of magnitude harder, since those data points are much rarer. As a result of the scarcity of historical data, the errors there are likely to be much greater.”

So - Gaussian Copula gave an analysis of risk that was unrealistic. Our banks used that forumla to take massive risks they didn't really understand on very illiquid products. In fact, the banks probably didn't even understand the massive risk because it was the Gaussian Copula that was used to rate the derivatives products triple A when they were junk.

Sure - Pablo's mortgage default may have been the last straw BUT it was the quantitative mathematics and the consequent massive risk taking that nearly brought down the financial markets.

 

[pedro01]

Quote from Champion:

Do you realise that newbie posters are responsible for a lot of poor thought and many garbage opinions brought into ET, yours especially included?

I haven't seen you make a contribution to the argument except that your post count is higher than mine.

Is ET one of these cliquey sites where it's just members with lots of posts stroking each other ?

How dissapointing...

 

[pedro01]

Quote from agpilot:

Quote from TraderZones:
I made $5K that day, by loading up on mutual funds at EOD prices. I figured it was a serious overreaction, and I cashed in.

--------------------------------------------------------------------------------

To TraderZone: I think your lying. Almost no body could get their orders filled that day. Big Fido quit answering their phones early that morning......... Post a copy of your P/L for "loading up" ... A whole $5k? Your big wad musta shook the market..
..
You make a good point Pedro01.. 15 post in 5 years but worth reading... ag

LOL - that's 3 posts a year! I think I need to back off a little - this interweb is addictive.

Actually - it's possible he made money that day - plenty of people must have been short from the day before - what a bonus if they were !

 

[hermit]

Quote from pedro01:

Dunno - I was only 17 then

So after 87, suddenly the crashes decided to act on what the quants did

 

[hermit]

Quote from pedro01:


The whole concept of the 'big boys' employing quants to write software that will predict the direction of the markets is a MYTH.

Direction has got nothing to do with it, you got your Elliott Waves and the Quant mixed up Sir.

 

[bobneedham]

Quote from HiddenAgenda:

"The average retail trader by definition does not stand a chance and the best advice to such a person is indeed: stop being average. A herd-following trader is doomed just like an artist without individuality." [/B]

I so agree with this! But you have to accept the fact that most new traders find it difficult to trade against the giants.

 

[pedro01]

Quote from hermit:

So after 87, suddenly the crashes decided to act on what the quants did

It would appear that you haven't read the posts but want to join in the debate for the hell of it.

In 87, complex mathematics turned what would have been a little crash into a bloodbath. Portfolio Insurance theory is a great concept BUT not if everyone does it at the same time.

If you read up on Portfolio Insurance (hell - it's in the above posts) and the avalanche of selling it caused, you would understand the role of quantitive analysis in the 23% one day drop.

 

[pedro01]

Quote from hermit:

Direction has got nothing to do with it, you got your Elliott Waves and the Quant mixed up Sir.

I don't think so - but perhaps instead of 'joining the crowd' on this thread, you could add some meat to the bone of your argument.

All I see on this thread is people taking a well formed analysis and saying 'no it isn't'.

Very few quants work on algorithms to predict the direction of the markets and producing automated trading systems based on those algorithms. This is the myth of the quant.

The reality of the quant is that they work on risk analysis, IT systems (e.g. HFT which is not predictive) and a host of other things not related at all to creating mechanical trading systems. Mostly they do stuff to help the real traders, although sometimes they produce eroneous risk models that almost bring the whole house down.

Of course, teh interweb is full of clever quants that work for Goldman Sachs on algorithmic programs....

... and I'm nailing Claudia Schiffer

 

[pedro01]

Quote from bobneedham:

I so agree with this! But you have to accept the fact that most new traders find it difficult to trade against the giants.

You don't need to trade 'against' anyone. The market isn't out to get any one participant however much it may feel that way. Some institutions (e.g. ETFs) have to buy/sell stocks REGARDLESS of the profit on the trade. There's a whole host of people trading in different directions for different reasons at any one time.

Stand in front of a herd of elephants and you will likely be trampled to death. Run after a herd of elephants and you may tread in the odd jumbo-turd but you will be alive and well.

Not knowing there is a herd of elephants behind you is the issue.

 

[hermit]

Quote from pedro01:

I don't think so - but perhaps instead of 'joining the crowd' on this thread, you could add some meat to the bone of your argument.

All I see on this thread is people taking a well formed analysis and saying 'no it isn't'.

Very few quants work on algorithms to predict the direction of the markets and producing automated trading systems based on those algorithms. This is the myth of the quant.

The reality of the quant is that they work on risk analysis, IT systems (e.g. HFT which is not predictive) and a host of other things not related at all to creating mechanical trading systems. Mostly they do stuff to help the real traders, although sometimes they produce eroneous risk models that almost bring the whole house down.

Of course, teh interweb is full of clever quants that work for Goldman Sachs on algorithmic programs....

... and I'm nailing Claudia Schiffer

Like everything else in the world, quantitative methods have their good and bad side and there are good quants and bad quants.

If there were no quantitative methods and we go back to trading through the pits, will the crashes and panics suddenly stop? What about all the scams/ripoffs that used to happen when everything was pit-traded? If you are seeking a utopia, markets are not the place for you.