How to start my education in Quantitative Analysis?

oxusmorouz · #11 18th April 2008, 02:38 PM

Quote:

Originally Posted by beginner_av

Guys its not that one day I can wake up and say hey let me become a doctor today even though I never studied Bio in 12th.

Rarely does one take a career decision overnight nor act on impulses for such issues.

Quote:

Originally Posted by beginner_av

Can you put hours practicing and doing QA instead of doing other things?

Maybe that is why I intend to take a full time systematic study?

dhakkan · #12 18th April 2008, 02:47 PM

my 2 cents !!!

Please add physics also in the list... though you might question why... I can not explain properly....but will try ....

In mathematics you device formula's etc..but the real implementation of these are in physics...in pure mathematical abstract world it sometime becomes very difficult to underastand.. what is what....
So a modelling of these mathematicall thories is required... physics helps there...

say for option .. we have delta and gamma...now gamma is rate of change of delta..

if you can visualize it ..great... but if not.. then Physics can help you ....

if delta is speed then gamma is acceleration....

makes it easy to visualize ..

beginner_av · #13 18th April 2008, 02:53 PM

Quote:

Originally Posted by oxusmorouz

Rarely does one take a career decision overnight nor act on impulses for such issues.

Maybe that is why I intend to take a full time systematic study?

You have started at the right age. Not many have that advantage. They are already doing something for years. I fully agree with your comment. Career decision is not an overnight choice, but neither is shifting careers that require complete different skill sets. Now would you have been better if you were a E&C enginner? Or will you be better by just picking up the required stuff from different places? Depends entirely upon you. But if you are looking for jobs, HR guys are very coconut headed. They wont even accept anyone's application without "proper" ABC letters against their names.
And why dont you take up a course in engineering maths as a starting point instead of everything else if you really want to become a quant?

oxusmorouz · #14 18th April 2008, 03:28 PM

Quote:

Originally Posted by beginner_av

Now would you have been better if you were a E&C enginner? Or will you be better by just picking up the required stuff from different places?

An engineer of course. That is why one has to undertake systematic education in any field.

Quote:

Originally Posted by beginner_av

And why dont you take up a course in engineering maths as a starting point instead of everything else if you really want to become a quant?

I seek to concentrate on the statistical approach over the mathematical one. Mathematical quants is not for those in the field of finance, but rather as Dhakkan pointed out, physicists, math PhDs and engineers. I would not fit into it. Let's see how the actuarial course unfolds.

beginner_av · #15 18th April 2008, 04:26 PM

Quote:

Originally Posted by oxusmorouz

I seek to concentrate on the statistical approach over the mathematical one. Mathematical quants is not for those in the field of finance, but rather as Dhakkan pointed out, physicists, math PhDs and engineers. I would not fit into it. Let's see how the actuarial course unfolds.

After CV, on this forum I can bet half my ranch on you. All the best and do keep us posted, and take this discussion forward.

As a FFT

What distribution for price do you use?

oxusmorouz · #16 18th April 2008, 05:56 PM

Quote:

Originally Posted by beginner_av

After CV, on this forum I can bet half my ranch on you.

*(laughs hysterically)

Quote:

Originally Posted by beginner_av

What distribution for price do you use?

Going with the assumption of normality in most cases, in spite its inherent limitations. As far as I can see, it is the safest of the distribution assumptions where data is not large enough.
If the data is large enough, and where feasible, I consider a histogram of historical data as a distribution in itself (especially in cases where prices are involved). You may say why not convert it into a alpha stable distribution instead? Just that if I do, I'll be adding one more assumption (not will will historical distribution hold good for immediate future, but will also have the same alpha) and the outcome will become more complex!

trader111 · #17 18th April 2008, 06:15 PM

Quote:

Originally Posted by beginner_av

You have started at the right age. Not many have that advantage. They are already doing something for years. I fully agree with your comment. Career decision is not an overnight choice, but neither is shifting careers that require complete different skill sets.

Why does one needs to be a quant after all? Is it the only way to make money in the markets? or the best way to make money from the markets?

But IMHO having an insight in the quant/algo industry is essentially required. As market participants we need to keep an eye on the 'edge' that other market participants are developing. That is all that matters.

Actually i was stupefied by members of the like aca_trader expressing their bias towards algo/quant traders. (here: http://www.traderji.com/trading-psyc...e-markets.html)

raosrinivas · #18 18th April 2008, 06:18 PM

Quote:

Originally Posted by trader111

Why does one needs to be a quant after all? Is it the only way to make money in the markets? or the best way to make money from the markets?

But IMHO having an inside in the quant/algo industry is essentially required. As market participants we need to keep an eye on the 'edge' that other market participants are developing. That is all that matters.

Actually i was stupefied by members of the like aca_trader expressing their bias towards algo/quant traders. (here: http://www.traderji.com/trading-psyc...e-markets.html)

well said, quants/ algo / techni / c++ really not necessary, of course you want to get into some job that's good. but traders make money using simple things .

beginner_av · #19 18th April 2008, 06:56 PM

Quote:

Originally Posted by trader111

Why does one needs to be a quant after all? Is it the only way to make money in the markets? or the best way to make money from the markets?

But IMHO having an insight in the quant/algo industry is essentially required. As market participants we need to keep an eye on the 'edge' that other market participants are developing. That is all that matters.

Actually i was stupefied by members of the like aca_trader expressing their bias towards algo/quant traders. (here: http://www.traderji.com/trading-psyc...e-markets.html)

let me tell u why. Quant is not a qualification, just a frame of mind. To not accept crap without scientific validity. Accept only that is verified. Some one, and we have lot of bullshi**ers around here, will come and tell you buy this as there is a 3 star afternoon white cloud, this is a 53mins 20 sec breakout etc etc. Now how do you verify it? Paste a chart and you will see 10 people naming 10 different formations. Cos the mind is used to seeing patterns when there is none. Some will see a wedge, some will extend it by 6 bars and see a triangle etc etc. But that doesn't happen if you can quantify your premise and verify it.

beginner_av · #20 18th April 2008, 07:04 PM

Quote:

Originally Posted by raosrinivas

well said, quants/ algo / techni / c++ really not necessary, of course you want to get into some job that's good. but traders make money using simple things .

nothing is necessary, but why dont you hand draw charts today or calculate PE ratio with pencil and paper? Because u want to take advantage of technology that can overcome human limitations. thats all.