Programming for Traders

Quote from arold_ite:

Somebody else mentioned that as this is such a specific area, the number of sales might not justify the effort and expense and it may be difficult to find a publisher willing to take on the risk. I’m kind of leaning towards this opinion myself. However, I guess there is always the self-publishing route.
More...

Put the manuscript together, lock down the rights to it, and then look at partnering with one of the major software vendors that has automatic trading as one of its components (like Ninja Trader), and/or retail brokerages (I'm sure Interactive Brokers would also be interested in such a venture as well).

This way you will have automatic access to their extended client base as a starter, and grow from there.

Alternatively, just put together a "program trading boot camp" manuscript and market it over the web for $300--$500 bills (for something on that level, it would have to have different modules which could take a trader to the proficiency level that he/she needs to be at to write their program and implemnt it ... nuts to bolts.


Best,

Jimmy
 
Quote from arold_ite:
In terms of languages, I will probably base it either on VB.Net or C#, mainly because these languages are easy to learn,
More...
Nice, i have two books about c# but it is so hard to get into programming and my goal is not to become a programmer i want only to some things. So i would also buy this book "Programming for Traders"! But i do not want to wait one or two years. What can we do now? You should do a wiki-website or something like this for the project: "Programming for Traders"...

If you really want to help people with practical and not with high intellectual abstractly stuff, you will become very famous AND rich! :)

Greetings from europe/germany
 
Quote from arold_ite:

I was talking with a trader friend this weekend and he suggested that I write a book for traders who want to learn how to program their own back-tests and trading systems, etc.

Anyway, I was wondering if anybody on this group has any opinions on whether there would be enough demand for such a book?

I’m thinking of aiming the book at the complete programming novice. Somebody that’s never programmed before but wants to gain an understanding of what’s involved before stepping up to the more serious programming resources. Of course, all of the examples and context of the book would be aimed at trading-specific solutions.

What do you think?

Arold
More...

I think this is a great idea! And I would be one of your first customers to buy your new book. It would have to be written for the complete novice involved. Since I am a complete beginner and never programmed anything, I can be a beta tester reader for your book if you need one. But I will buy your book for sure. Good idea.
 
Quote from biologymajor:

I think this is a great idea! And I would be one of your first customers to buy your new book. It would have to be written for the complete novice involved. Since I am a complete beginner and never programmed anything, I can be a beta tester reader for your book if you need one. But I will buy your book for sure. Good idea.
More...

A lot of traders would be behind you if you gave it a serious effort. (I mean, just think of all the guys with a full-time job, with the time invested education and/or familial obligations, who have a viable strategy, but they just don't have the time to implement it).

Me, I'm free as a bird and like that way, but heck, I'd definitely be one of your beta testers!

Keep us abreast.

Best,

Jimmy
 
This sounds like a good idea in theory but I am skeptical that it has any value in practice.

1) Learning to program has <i>nothing</i> to do with trading. If everything you need to know to write a useful backtest engine or automated trading system is a thousand pages of information, about fifty of those pages concern finance or financial instruments, and the other 950 are about programming, software engineering, technology, probability and statistics. I don't see how a trader-specific programming text adds much value for the vast majority of information you need to impart.

2) For a non-programmer who wants to backtest or trade automatically, learning a general purpose programming language is probably not the best route. Excel, Matlab, Wealth Lab, EasyLanguage, etc. are a lot more plausible. I doubt there's a single person reading this thread who wrote their own automated trading system from scratch without considerable prior experience as a software developer.

3) As a corrolary, if you really want to enable novices to backtest and trade mechanical strategies, instead of writing a book about programming, you should write and sell a system like EasyLanguage or WealthLab.

4) A new programmer should not write a backtest engine or automated trading system as their first project. A good first project is something self validating where errors are immediately obvious and test cases are easy to come by. Backtesters can fail silently for subtle reasons, and in doing so appear to have uncovered the key to fabulous wealth. And automated trading systems - well that should be obvious. Might as well start a novice programmer writing software for a surgical robot, it's about as prudent.

5) Finally, talk to 10 different successful automated traders and you might hear about 5 programming languages, 3 operating systems, 8 execution and market data APIs, 6 database backends, and probably a dozen financial markets. If you describe just one of them you are doing your readers a disservice. If you talk about all of them you have a 3000 page book.

Martin
 
I've been wishing for something like this for awhile now. I was thinking of starting a blog or something to help people out (and myself :p) on this road, since there is nothing out there right now. The problem for novices is that trying to learn a programming language or programming concepts like OO through reading a book and then transferring those concepts into a trading app is a huge jump. I ended up taking classes at the local JC and had tutors help me build small projects that I wanted designed. Even now though I feel like programming a framework is a massive undertaking. If something was there that helped someone down the right path would be amazing. I would be a definite buyer.

Chicago
 
Quote from Sparohok:
This sounds like a good idea in theory but I am skeptical that it has any value in practice.

2) For a non-programmer who wants to backtest or trade automatically, learning a general purpose programming language is probably not the best route. Excel, Matlab, Wealth Lab, EasyLanguage, etc. are a lot more plausible.
More...
Excuse me Martin (Sparohok), but what you wrote is a theory. I tried months to connect excel to IBs TWS. And than if i have a stable activeX connection, the stumpiest things dont work. An example for a stupid "Times&Sale"clone:
Code: 
Private Sub Worksheet_Change(ByVal Target As Range)
   If Not Intersect(Target, Range("D2:E2")) Is Nothing Then 'watched area
      Application.EnableEvents = False
      Range("A2").Value = Date
      Range("B2").Value = Time
      Range("B4").EntireRow.Insert Shift:=xlDown
      Range("A2:H2").Copy
      Range("A4:H4").PasteSpecial Paste:=xlPasteAll
      Application.EnableEvents = True
      Application.CutCopyMode = False
   End If
End Sub
This Worksheet_Change is not working if a price or size is changing in a cell! Why? This is everything but not plausible! And i can do dozen of posts in Forums and nothing is going forward...
But, i think also it is only possible by concentrating of one broker api and go with practical examples through "all" importend steps. Like a tutorial for IB, because there are all three requirements via api:
- realtime data
- historical data
- order execution
 
excel was just one of the options he mentioned bro, and the IB API isn't the only one in town

i really believe that is possible to create a consistently profitable, automated system

first you have to be a really good trader

then you have to be a really good programmer

:D :)

so therein lies the difficulty

it's difficult to be one OR the other

it's extremely difficult to be both (think of it as having a double Master or Ph.D degree)

later,

jimmy
 
You must log in or register to reply here.
Back
Top