C++, Visual C++ 6, Visual C++.net, C#

Unfortunately EdgeHunter is 100%. You will be shocked for what VB/Excel-VBA are used for, in terms of lets-test-this-quickly in the 'City' or the 'Street'.

With regards to Quants, I am not aware of any wholesale firing - unless that means a good cleansing of the deadbeats - and this is constantly happening in firms that need quants. Do you have any links to this?

With respect to what they do, it never ceases to amaze me how many here really have no idea...the following is what I posted a while back in another thread

I believe that there is a bit of confusion when it comes to Maths/Engineering/Physics/Finance PhDs and the kind of work that they do.

The vast majority of Quants develop, implement (i.e. code), validate, calibrate, and maintain pricing, risk, and hedging models for derivatives, sometimes rather complex derivatives - the sort of shit that make s you go grey prematurely. Have a look at some of the job adverts on Quant websites, and you'll see what the requirements are.

Yes, there are Quants working on algorithmic trading, but a lot of the work done here is to reduce execution risk and market impact of trading strategies (i.e. for crude examples - VWAP or TWAP transactions of huge orders, etc, etc).

Of course, there are also bots making markets and an increasing number of Quants working on fully automated trading.

How many guys here that appear to have some knowledge in algo-trading even know what a CDS is? Of course, why should they?

The point is that there are many facets to Quant work, from grunt quant development, to the elite model development, to financial engineering, risk management, effective hedging, etc, etc...
 
Quote from Equalizer:

...With respect to what they do, it never ceases to amaze me how many here really have no idea...the following is what I posted a while back in another thread...
More...
It doesn't surprise me one bit. Lots of people have opinions on this site on subject matter they have zero idea about. I have noticed that seems to be the human condition especially on the Internet but in the real world as well.

nitro
 
Quote from mahras2:

There are quantitative funds making multiples of what Livermore made in his lifetime. And this is without blowing up multiple times, filing bankruptcy and committing suicide. And they have a better life to boot.
More...
Kind of you to tell us. Fill us in also on whether they prefer C++ or go rather for .NET stuff. That's what the man wanted to know.

I'll add that in Livermore's days there were guys making in Livermore's-days-$ what he made without blowing up multiple times.

:D
 
Quote from Equalizer:

...
How many guys here that appear to have some knowledge in algo-trading...
More...
I leave this up to you to figure out, Equalizer.
All what I can say is that after reading everything in this thread about them supposed 'quants', I got a terrible nononsense itch.

:cool:
 
I think there are a lot of mis-information here. First of all, it is clear that there are both technologists and some analysts in this thread. I have been both (but primarily as a technologist).

The real answer is again one of those "it depends" type of answers. it depends what the application is designed to do. For instance, if an ibank wants to sell a custom call option for a custom basket of equities, using VB and Excel would be sufficient with interfaces with a centralized theoretical engine. Naturally, the continuous central theoretical engine would be written in C++, for speed and efficiency. Heck, I know a few hackers in wall street that would refuse even to use C++ (but would use raw C or assembler), given there are noticible performance gains by going to a lower-level language (roughly 7-10%). But that's besides the point.

Most quants I know would use VB if it is one of those "one-off" type of analysis and prototype. However, this turns into a mess pretty quickly, as those "one-off" type of prototypes may proven to be good and popular, and become an application all by it self. Statistical languages (i.e., S-Plus) have also a lot of popularity in the Quant community.

Yes, if you were to write an automated trading system or something that requires real-time data analysis, you would be dumb not to use a compiled language (C/C++ being the prominent one, but there are some die-hard NeXT users that use Objective-C).

On the topic of quants be fired "amass", that would be news to me, most of the ibanks I know are having enormous trouble hiring good Quants. I ran a small hedge fund analytics group back in 2001-2002 for an ibank, all of the people that used to be in my group are now calling around looking for people (most of them run their own group now). A friend of mine that runs a MMF (Masters of Mathematical Finance) program received record application (26x the number of student that would be accepted) this last year.

However, the pure theoretical Quants are now looked at with a bit of suspicion. Ibanks used to swoon over each academic hire (back in early 90s), they now require they spend a year or two actually modeling or trading before given them free reign. A fellow I know who used to be associate professor of Finance (in a tier-one univ, with good publications) was hired 4-5 months as a Director (only) of quant strategies, when a MD position would be the norm just 3-4 years ago.

Amaranth have not had a good year performance wise (look up, yep, their return for the first 6 months of 2005 is around 0.5%). They grew very fast, from only <$2B in 2002 I believe to now 6-7B. I think they are experiencing the limitations of their model. They have always been very tech driven, they were looking for a head of arbitrage systems about 3-4 months ago, not sure who they ended up grabbing.

Quote from prt_systems:

For the largest bond fund we used/use only C++ - not $soft. However some of the systems were based upon the $net/$soft platform but the majority of the mission critical systems that actually run the funds were based upon straight C++ or other systems. The $soft stuff was used on some of the front end systems. In my own operations I used the designs I did for the fund and re-architected portions of hte design - better than the integrations I did at the bond fund - and moved that into strictly open source solutions - with some memory resident databases or the capability to turn this on for heavier loads.
More...
 
i think many of us have experienced something like this before. the point really drove home for me when i read a chapter from the 48 Laws of Power, by Robert Greene.

Law 7: Get others to do the work for you, but always take the credit.

"There are those who live by hunting and killing, and there are also vast numbers of creatures (hyenas, vultures) who live off the hunting of others. These latter, less imaginative types are often incapable of doing the work that is essential for the creation of power. They understand early on, though, that if they wait long enough, they can always find another animal to do the work for them. Do not be naive: At this very moment, while you are slaving away on some project, there are vultures circling above trying to figure out a way to survive and even thrive off your creativity. It is useless to complain about this, or to wear yourself ragged with bitterness, as Tesla did...

This is the essense of the Law: Learn to get others to do the work for you while you take the credit, and you appear to be of godlike strength and power. If you think it important to do all the work yourself, you will never get far, and you will suffer the fate of the Balboas and Teslas of the world. Find people with the skills and creativity you lack. Either hire them, while putting your own name on top of theirs, or find a way to take their work and make it your own. Their creativity thus becomes yours, and you seem a genius to the world."




Quote from prt_systems:

This actually reads: We want someone with a PhD and ten years of hard core experience, whose systems have made 100's of millions of dollars and who will work 70 hours per week.

We will pay 130K with employment at will: Once we are done with your building our systems will we let you move on and your revenue stream is over while we make our 100 million using the technology.


...OR you can just use your skills to build your own company. Hmmm I wonder which is the better choice ?
More...
 
I have never seriously coded in VB (other one or two Excel basic toys), so I have no idea there.

For C++, the basic language introduction will take 1-2 courses by itself. But assuming you have knowledge of another OO language and have some knoweldge of programming languages in general, you can pick it up fairly quickly. I have seen somebody picked it up in 2 weeks. However, you will need some practical experience before you can get some idea of the tips and tricks, especially the C generated structures. To get some handle on the C++ libraries (network programming, threading, job scheduling, etc), well, that's a life-time of work. I started with C++ in '87 (ANSCI C++ v2, no templates), worked on porting some X11R5 to C++, and still I don't know all the little architecture-dependent tricks.

A good place to start to high-performance C++ is actually graphics and maybe crypto libraries (BN library anyone?). In my opinion, the performance tuning and compact design in finance doesn't hold a candle to all the crazy code tweaking and optimization that goes on in graphics engines. Of course, Carmack codes mostly in C (and un-elegant at that), but it works, and it is blazingly fast. In my opinion, there are too many class designers in finance systems.

Quote from jerryz:

how many C++ and visual basic courses would you guys say are sufficient?
More...
 
Quote from rufus_4000:

I think there are a lot of mis-information here. ...
More...
There sure is Bro! But what did you expect? This is ET where quants are not only misunderstood but perhaps even ridiculed. Then again look at the treatment chartists get at the Wilmott forum. In between the two extremes there is the truth.

Can't we all just get along :D
 
Quote from Equalizer:


Can't we all just get along :D
More...

Not if you use Java... :eek: :D

:)

cj...

__________________
HAVE STOP - WILL TRADE

If You Have The Vision We Have The Code
 
You must log in or register to reply here.
Back
Top