 |
SteveH
Registered: Mar 2004
Posts: 146 |
07-12-12 03:53 PM
It probably depends more on your programming ability. If you've got those skills, then you'll always find (at some point) something in TradeLink or NinjaTrader, etc which you feel is a hindrance to expressing what you want to do. The biggest advantage of roll-your-own is that you don't have to please anyone else. You can go deep where you want to and fix bugs extremely fast.
However, I do think the main developer of TradeLink has done a great job of moving his project forward. It is one of the best free resources for a multiple broker trading interface.
|
| |
|
Edit/Delete • Quote • Complain |
bluelou
Registered: Nov 2006
Posts: 190 |
07-12-12 04:11 PM
SteveH,
Yes, very true. I'm not enough of a programmer to mess w/much other than the strategy itself.
I'm sure TradeLink is a fine product. I just need to come to terms with the pluses and minuses of an open-source platform.
|
| |
|
Edit/Delete • Quote • Complain |
amazingIndustry
Registered: Aug 2009
Posts: 570 |
07-13-12 01:09 AM
I actually used RightEdge long time ago purely for feeding a simple strategy engine with historical data. But I hit the limits very soon and that was when I branched out and started writing my own platform. RightEdge is not bad for a start but I never trusted the statistics of any of those off-the-shelf products. I ended up writing all output statistics and trades and analyzed myself. In the end I wrote more and more parts myself so that I did not see the purpose of using RightEdge anymore. My main gripe was that I run multi-symbol backtest feeds and RightEdge had a very hard time handling those. If you just stream one single symbol's time series into your NewTick function then thats fine (even then slow) but doing so over 10 symbols slowed things down immensely.
Quote from bluelou:
I understand. Given the opacity you've written of I don't have any interest in trying to tease out answers to "Where do I find/How do I do this..." questions to a paid email help service. Developing and maintaining strategies is hard enough. But, I'll still continue to monitor what's going on at Tradelink.
Did you ever consider RightEdge? If so, what did you like/dislike?
While I know that Ninja isn't suitable for many, in combo w/R and some statistical libraries attached as DLLs it's okay for me for now. I spoke to them recently and they mentioned adding portfolio trading and a cloud-based service but I have no idea of time frame.
Thx again,
Lou
|
| |
|
Edit/Delete • Quote • Complain |
amazingIndustry
Registered: Aug 2009
Posts: 570 |
07-13-12 01:15 AM
1) How true is that: Yes I also believe its one of the biggest advantages, should one hit a bug you can just debug it yourself because you understand your own code the best. It took me hours to figure out the mechanics and search for a solution to errors in products, written by others. You sometimes end up waiting for days for someone to help out on message boards. Thats just the nature of working with stuff that is not your own.
2) Agree, at least targeting .Net. There are couple other products targeting Java that I would claim are superior to Tradelink. MarketCetera comes to mind and another which I looked at some time ago but which name slipped my mind. But the Tradelink guys did a great job considering its open source and targeting .Net at a time .Net really was not in favor at all by the financial community. The biggest problem, as outlined before, is that non-Tradelink-related takers of Tradelink are very rare, its pretty much fully extended and supported by the very same guys who came up with the product originally. When you have a highly successful open source project you see various people coming in and out and committing new code, expanding ideas, testing,.. this I do not see is the case with Tradelink, at least not to the extend I would feel comfortable embracing this product.
Quote from SteveH:
The biggest advantage of roll-your-own is that you don't have to please anyone else. You can go deep where you want to and fix bugs extremely fast.
However, I do think the main developer of TradeLink has done a great job of moving his project forward. It is one of the best free resources for a multiple broker trading interface. [/B]
|
| |
|
Edit/Delete • Quote • Complain |
bluelou
Registered: Nov 2006
Posts: 190 |
07-13-12 01:28 AM
Quote from amazingIndustry:
I actually used RightEdge long time ago purely for feeding a simple strategy engine with historical data. But I hit the limits very soon and that was when I branched out and started writing my own platform. RightEdge is not bad for a start but I never trusted the statistics of any of those off-the-shelf products. I ended up writing all output statistics and trades and analyzed myself. In the end I wrote more and more parts myself so that I did not see the purpose of using RightEdge anymore. My main gripe was that I run multi-symbol backtest feeds and RightEdge had a very hard time handling those. If you just stream one single symbol's time series into your NewTick function then thats fine (even then slow) but doing so over 10 symbols slowed things down immensely.
AI,
Were you running multi-symbol backtests on tick data or time bars? If I could run portfolio/multi-symbol backtests on time bars w/RightEdge and it wasn't unbearably slow I could live with that. Would that work with RightEdge? If it wasn't for that issue would you recommend the product, say, vs. Ninja?
I agree, those stats are dangerous. Besides, what conclusion am I supposed to draw from a strategy run on a single realization of a time series? LOL.
|
| |
|
Edit/Delete • Quote • Complain |
amazingIndustry
Registered: Aug 2009
Posts: 570 |
07-13-12 01:48 AM
it comes down to the same thing. For one to feel comfortable with a back test one has to understand the law of large numbers and its convergence properties, thoroughly understand the underlying statistics and probability theory when analyzing results. But above all I would say you need to operate over sufficiently long data sets which cover different dynamics, environments, market cycles. It always costs more to build bars than simply passing on loaded tick data. Running RightEdge over millions of tick data is slow. Running RightEdge over 1 or 2 years of hourly bars or even a year of minute bars is doable. Are they the same? No, think about how many observations hourly bars over a year get you vs. a year worth of tick data...I guess you get my point. RightEdge is just not really fast iterating over data. That the package is not really optimized for professional users shows that it offers a Jet and MSSQL adapter to load data from such databases.
Quote from bluelou:
AI,
Were you running multi-symbol backtests on tick data or time bars? If I could run portfolio/multi-symbol backtests on time bars w/RightEdge and it wasn't unbearably slow I could live with that. Would that work with RightEdge? If it wasn't for that issue would you recommend the product, say, vs. Ninja?
I agree, those stats are dangerous. Besides, what conclusion am I supposed to draw from a strategy run on a single realization of a time series? LOL.
|
| |
|
Edit/Delete • Quote • Complain |
| Receive
an email whenever a new post is added to this thread by subscribing
to it. |
|
|
|
|
