The Birth of a Quant Trader | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Quant basket trader? I need your help! 92 replies

Quant trading & QuantLib 6 replies

SAS for quant analysis 0 replies

Quant Corner 12 replies

questions from a future quant 39 replies

The Birth of a Quant Trader

  • #1
  • First Post: Mar 20, 2011 2:27pm Mar 20, 2011 2:27pm
After a lot of thought, soul searching, research and plain reality, I have concluded that my time and energy is going to be focused exclusively on becoming a quant trader and on developing multiple quant systems.

Simply because I now believe in clear objective thinking, rather than wishful subjective thinking by having a robust time proven edge FIRST, and then a good psychology/discipline.

And for me this is through quant/automated systems.

However, my background isnt in programming, math, science, physics or anything as such, so I know theres a long path ahead of me.

Considering this is something new to me and that there is a lot of BS online, id like to humbly ask for some advice and recommendations with regards to information I must read and learn and if possible any good sources of information which I can gather the info from:

- Programming software to automate and test ideas.
- Research papers with regards to volatility and/or any 'important' data.
- 'Must know' keywords of commonly used technicalities by quants.
- Specific subjects required to become a quant.
- Suggested reading material with regards to quants/markets/programming.

Pretty much any and all info required to put me on the long and narrow road.

Those that know me know I dont look for hand holding. Just some direction...

My aim? To build multiple systems that complement each other under all market conditions.

I know what I want, I know that sooner or later I'll get what I want but im not sure which path (knowledge wise) I must follow.

If ive said anything silly (which i probably have ) please forgive my ignorance.

Im all for learning in a humble manner

All I ask is for the topic to keep on track.
  • #2
  • Mar 20, 2011 2:33pm Mar 20, 2011 2:33pm
Have you read "The Black Swan"?
  • #3
  • Mar 20, 2011 2:52pm Mar 20, 2011 2:52pm
Read any thread on http://www.nuclearphynance.com/
It will all sound like gobbledygook, but at least you will get some terms/areas to search on. It will also give you a sense of the gulf in knowledge between here and there.
The breaking of a wave cannot explain the whole sea.
  • #4
  • Mar 20, 2011 5:05pm Mar 20, 2011 5:05pm
Quoting Pipalicious
However, my background isn't in programming, math, science, physics or anything as such, so I know there's a long path ahead of me.
How long do you believe the path to be?

I say that it will take you at least 2 years of 8 hour/day to get the basic programming/quant knowledge required to be able to implement a minimal system. But you'll need a lot more to be able to be profitable.

And much longer if you'll do this in your free time.

Not knowing either programming or math/statistics puts you at a huge disadvantage for what you want to accomplish. It's hard enough if you only know one part of the equation (programming or math). That's why banks or hedge funds have quants doing research and programmers converting the prototype implementation delivered by the quants to a robust one suitable for actual use.

Now, back to your question, I used this approach to get papers: I searched for stuff like "predicting the forex market" and downloaded the search result PDFs which looked like academic (they have a very specific formatting style, you'll learn it quickly). Then I read through the papers, and noticed what terms it used, like "deseasonalization", or "volatility clustering", or "principal component analysis". Then I took these terms and searched for papers on them, and then looked through these papers and found a new set of terms to search for. In this way, I quickly learned the basic language that quants use. I still find it hard to read the equations, but at least I have a basic idea now of what the paper speaks of and what their method is.

Then I found out about Google Scholar, which searches only academic papers. Here's a starting point: http://scholar.google.com/scholar?q=forex

Another track is reading interviews with quants or with people from automatic trading firms. You can get some really nice insights in this way - for example that most automatic trading systems have success ratios of 45-55% (with avg win bigger than avg loss) , not >90% (with avg win much smaller than avg loss) like most retail traders try to build.

I think that everybody should do what it's good at, and that you could be seriously underestimating what it's required to do a WORKING automated trading system.
  • #5
  • Mar 20, 2011 5:41pm Mar 20, 2011 5:41pm
I came across http://quantlabs.net/ the other day.

They have a lot of research papers etc on the site.

Of course, the majority of them looked like they required the intelligence of an advanced alien life form, but hey have a look
Drink Coffee. Do stupid things faster with more energy.
  • #6
  • Mar 20, 2011 6:09pm Mar 20, 2011 6:09pm
read what you want and understand what is going on and you probably do already but '' they '' are writing new ones everyday
  • #7
  • Mar 21, 2011 3:59am Mar 21, 2011 3:59am
yep I have a long path ahead of me but I dont mind at all im in this for the long term. No rush....

Im already onto some solid ideas but the downside is the programming aspect of course. Hence why instead of 'hoping and wishing' relying on others that could pretty much just walk of with my research id rather learn it myself.
But im open to partnership.

As for 'them' continually coming out with new systems thats understandable.
However the markets do present exploitable opportunities if one is to research the 'correct' content.

Thanks for the advice on research that is pretty much what I do (google, search, research specific words/phrases, understand them, rinse and repeat).

As for automated systems they dont neccesarily have to be fully automated.

I also believe that many 'retail' automated systems fail simply because they are not based on truths that the market present and based more on subjective interpretations such as conventional technical analysis. But hey, that is my opinion. Not here to argue it.

As for black swan I havent read it. I always get somewhat skeptical by 'entertainment' books. Im not saying it hasnt got any value, in fact a friend of mine recommended 'fooled by randomness' over the black swan.

Either way im much more inclined to read up on research/academic content rather than 'best selling books'. Again, this is my opinion. I know theres always some nuggets to pick up on in these type of books (many which I have read...).

Heres another question if you guys dont mind....

'What where the reasons behind you becoming a quant as opposed to a conventional chartist?'

cheers!
  • #8
  • Mar 21, 2011 5:58am Mar 21, 2011 5:58am
Hello,

First thing you need a strong backtesting platform. MT just doesn't cut it...
You have a multitude of choices: build one yourself (time waster for me) or buy one (Amibroker, Tradingblox, TradesStudio etc). Or free ones although I don't have experience with any free solution other than MT.

Good luck.
Trading = a mirror to your human flaws. Fix them or be fixed.
  • #9
  • Edited 6:13am Mar 21, 2011 6:00am | Edited 6:13am
Quoting Pipalicious
After a lot of thought, soul searching, research and plain reality, I have concluded that my time and energy is going to be focused exclusively on becoming a quant trader and on developing multiple quant systems.

Simply because I now believe in clear objective thinking, rather than wishful subjective thinking by having a robust time proven edge FIRST, and then a good psychology/discipline.

And for me this is through quant/automated systems.
Happy to be of help if I can, but to begin with I'd like to question your premises.

What is the basis for your hypothesis that clear objective thinking requires substantial investments in quantitative and programming skills?
  • #10
  • Mar 21, 2011 6:53am Mar 21, 2011 6:53am
Quoting Pipalicious
Either way im much more inclined to read up on research/academic content rather than 'best selling books'. Again, this is my opinion. I know theres always some nuggets to pick up on in these type of books (many which I have read...).
I am fairly familiar with the academic literature on this, and you won't find your "robust time proven edge" there. Rather, you'll learn how to thoroughly test trading rules, and then see those methods applied to fairly simple trading rules, such as MAs, breakouts, and so on.

Quote
Heres another question if you guys dont mind....

'What where the reasons behind you becoming a quant as opposed to a conventional chartist?'

cheers!

I've got the skills and experience required, but I don't think they are of much use for a "robust" trading edge. Much more important is understanding the market. So read up on market micro-structure and its implications for trading opportunities. See the papers by Carol Osler for a start.
  • #11
  • Mar 21, 2011 7:16am Mar 21, 2011 7:16am
Quoting Pipalicious
yep I have a long path ahead of me but I dont mind at all im in this for the long term. No rush....
Consider the possibility that your road may be infinitely long. Read what one brilliant educator, Underwood Dudley, said about learning mathematics:

The proof may seem to be unsatisfying: each step is correct, and hence the conclusion is true, but it is not clear why the steps are there and where they came from. That is because there are at least eight levels of mathematical understanding, and it is hard for someone on a lower level to appreciate what goes on at a higher level. The levels are, I think:
1. Being able to do arithmetic.

2. Being able to substitute numbers in formulas.

3. Given formulas, being able to get other formulas.

4. Being able to understand the hypotheses and conclusions of theorems.

5. Being able to understand the proofs of theorems, step by step.

6. Being able to _really_ understand the proofs of theorems: that is, seeing why the proof is as it is, and comprehending the inwardness of the theorem and its relation to other theorems.

7. Being able to generalize and extend theorems.

8. Being able to see new relationships and discover and prove entirely new theorems.

Those of us stuck on level 5 can no more understand the workings of a level 8 mind than a cow could understand calculus.

(This is from his Elementary Number Theory)

To independently develop a robust quantitative trading system you would need skills at level 8.
  • #12
  • Mar 21, 2011 8:55am Mar 21, 2011 8:55am
I agree.

Ive been recommended Multicharts which im currently testing out and its really really good IMHO. But slightly out of my reach as of right now (€).

I also found this: http://www.smartquant.com/ which looks pretty solid aswell.

Someone else recommended me ninjatrader.

Obviously all this depends whos hands they are in...

Quoting soso
Hello,

First thing you need a strong backtesting platform. MT just doesn't cut it...
You have a multitude of choices: build one yourself (time waster for me) or buy one (Amibroker, Tradingblox, TradesStudio etc). Or free ones although I don't have experience with any free solution other than MT.

Good luck.
  • #13
  • Mar 21, 2011 9:10am Mar 21, 2011 9:10am
It isnt that 'clear objective thinking = programming/quant trading' as such...

What I mean is that conventional TA (IMHO of course...) can be very subjective based on who and how someone looks at the charts.

One can look at the charts, do their technical analysis (s/r, TL's, etc....top down approach....) and once finished, DELETE all their analysis, start from scratch again and ironically will (potentially) see something similar if not different to their FIRST analysis. (All this from personal experience of course).

So in essence to me so far it seems very subjective and dependent on the persons interpretation. Where is the consistency here?

If we all do this, then IMHO it creates complete chaos because we all see things in a different way.

So when I say 'clear objective thinking' what im trying to say is that under all this subjectivity/chaos there is certain repetitive 'events/patterns/behaviors' that can be quantified (to such extent) to the point that it is the numbers/patterns that speak and not so much my own interpretation of the charts.

Its clear that traders can trade without charts. So if they are trading without charts then they arent particularly doing conventional TA. And if so, what are they doing?

This of course is from my current belief system. I could be completely wrong. In fact, I very likely will be....but it does make sense to me personally so far.

Then there is the typical "be disciplined", "stick to your trading plan", "develop your mindset"...

Which I think is great and neccesary of course!

However, isnt this the same as simply 'quantifying' a system to such extent that it "becomes" the trader itself? 100% discipline, 100% objectivity (since no emotions involved...), solid mindset.

The trader itself becomes the risk manager, the 'system' the action taker.

I seriously believe that most automated systems are built with the wrong data. Most (IMHO) auto systems are built on conventional TA tools.

If we eliminate all tools from our charts, and look at them in plain vanilla, there is 2 specific variables that can be used to build systems with, be it fully automated or simple 'proprietary' indies.

I thank you for asking man because your making me think

And I dont intend to argue with anyone, just have an honest humble discussion. Id love to see and hear from other peoples perspectives.

In the end, aint that how we learn?


Quoting medici
Happy to be of help if I can, but to begin with I'd like to question your premises.

What is the basis for your hypothesis that clear objective thinking requires substantial investments in quantitative and programming skills?
  • #14
  • Mar 21, 2011 9:18am Mar 21, 2011 9:18am
Quoting medici
I am fairly familiar with the academic literature on this, and you won't find your...
I appreciate your advice on micro structure in fact Oslers papers are on my to-do list.

I just find it hard to organize and prioritize all my research

In the end, I think its ironic how one begins a voyage that takes you deep into the rabbit hole, only to come back out and find a simple little rabbit hidden in it.

I.e. partake on complex research, trial and error, put simply....curiosity simply to come back with a "less is more" approach or a "keep it simple" mentality.

My aim with all this is to keep it simple and straight to the point.

I dont want to become some nuclear physicist who can build rockets in his spare time. I value my time immensely as you do as well.

But it is something I MUST research to get of my neck the "what if..." splinter.

The more I think, the more I realize that to build a solid future in this 'volatile' industry is by developing systems (be them manual or whatever) that complement each other under all market conditions...

Thanks for the heads up on Osler
  • #15
  • Edited 9:51am Mar 21, 2011 9:20am | Edited 9:51am
Quoting medici
Consider the possibility that your road may be infinitely long. Read what one brilliant educator, Underwood Dudley, said about learning mathematics:

The proof may seem to be unsatisfying: each step is correct, and hence the conclusion is true, but it is not clear why the steps are there and where they came from. That is because there are at least eight levels of mathematical understanding, and it is hard for someone on a lower level to appreciate what goes on at a higher level. The levels are, I think:
1. ...
Nice list.
In other words, be able to do calculus in your head.

(I think I was at 3 or 4, since unused, if about reading chart patterns a bit better.)
  • #16
  • Mar 21, 2011 9:30am Mar 21, 2011 9:30am
Quoting medici
To independently develop a robust quantitative trading system you would need skills at level 8.

What do you suggest whole heartedly? What do you recomend I do?

What would you do if (knowing everything you know now) you could start from scratch? day 1...

Im no genius. I also believe one doesnt have to be to achieve what he wants from life (so long...as it is what one truly wants at a deeper level.)

I believe simple systems, based on simple measurable data can be built to succeed with. Without the need to dig deeper into quant finance.

I also wonder,

If we take psychology, mindset, lack of confidence, greed/fear out of the equation...

And only have conventional systems that people trade now publicly such as this forum.

Why do most people continue to struggle?

Why do most people keep failing no matter what?

Is it because the systems themselves are floored? Over exploited?

Could it be with the subjectivity involved in TA?

Not to argue man, im just openly sharing what crosses my mind.

I probably sound very ignorant. IN FACT! I bet (no pun intended) that someone will read this and think...'hmm, hes got a hell of a ride yet!'
  • #17
  • Mar 21, 2011 11:25am Mar 21, 2011 11:25am
I don't think it's worth the time.

Not that you are trying to be a quant trader on the street or anything, but just to put some perspective on things, these jobs generally require a graduate level understanding of maths/stats, on top of programming skills/experience.

I'd venture to guess you could learn the programming on your own time, if given enough time. I wouldn't feel comfortable making the assumption that you could learn the math that phds in physics are using on your own (but of course anything is possible).
  • #18
  • Mar 21, 2011 12:03pm Mar 21, 2011 12:03pm
Quoting smittens4212
I don't think it's worth the time.
...
I disagree. You don't need to have a graduate level understanding in maths/stats (if graduate level understanding is what I assume it is).

You need to have basic understanding of those to start it up. Have a look at http://www.automated-trading-system.com/, "simple" trend following strategies still work albeit not like back in the 80s when they were generating hundreds of % profit per year. But it's a good starting point.

Programming is a must though. To be able to test your ideas.
Trading = a mirror to your human flaws. Fix them or be fixed.
  • #19
  • Mar 21, 2011 12:14pm Mar 21, 2011 12:14pm
Quoting Pipalicious
What I mean is that conventional TA (IMHO of course...) can be very subjective based on who and how someone looks at the charts.
...
So in essence to me so far it seems very subjective and dependent on the persons interpretation. Where is the consistency here?
Trading is about more than just TA (and I don't mean fundamentals): if you and I both look at a chart, you say price is about to go up, and I say price is about to go down, is one of us necessarily wrong?
No. (btw, we both may be wrong - or we both may be right).
The consistency is in how we trade these setups. Perhaps on a short TF price eventually increases, but on a longer TF it decreases. If you take your profits during that short timeframe, but I hang on, we may both make profits. If we are beginners, the situation may be reversed, and we both lose.
More importantly, don't forget that this whole discussion should be viewed in a statistical context: pro's also have many losing trades, but overall they're profitable. That's were the consistency comes in: trading, not merely TA.

Quoting Pipalicious
I seriously believe that most automated systems are built with the wrong data.
I'm not sure I understand where this belief is coming from. Do you mean wrong indicators? Do you mean wrong parameters? Or perhaps you meant backtested with the wrong data? What's the right data? Please clarify.

Quoting Pipalicious
Most (IMHO) auto systems are built on conventional TA tools.
And what's wrong with that?
If you ask me, it's not that the "conventional TA tools" are the wrong tools. In fact, I think they're great!
However, too many traders misuse them. As I see it, these tools are supposed to be used to quantify certain price patterns, and react, for example, when certain thresholds are crossed (Keep in mind that many of these conventional tools were developed many years (or even decades) ago, before flashy colorful MT4 terminals took over our CPUs). All too often what you see on this forum, is a trading system that almost blindly chooses an indicator, without understanding what it measures, because a backtest (of limited scope) provided favorable results [for anyone who's having success with this: may your success prevail].
So you see, it's not that right tools are not there - it that we've come to accept their misuse as natural.

Good luck!
And consider reading www.ieor.columbia.edu/pdf-files/Paul_Dominic.pdf
  • #20
  • Mar 21, 2011 2:45pm Mar 21, 2011 2:45pm
Quoting 6pack
So you see, it's not that right tools are not there - it that we've come to accept their misuse as natural.

Good luck!
And consider reading www.ieor.columbia.edu/pdf-files/Paul_Dominic.pdf
One of the most intresting statements I have seen on this site.
2 traders viewing now