Search results

Hi Mhtrader, I admit I didn't read the post I linked with prejudice. But I see neither did you read it very thoroughly. The OP in http://www.daniweb.com/forums/thread155136.html tests C++ code against his own assembly code "inlined" in a C++ function and the same assembly code compiled as a...

Out of curiosity, what was the motivation to use gcc?

Haha! Thanks for the fun link Tom, here you are addressing the real issue. "The best" language is usually the one the programmer is most familiar with. Hi NetTecture, my post you quoted was actually addressing specific points Fundjunkie made. Funny enough, the discussion I linked earlier...

Are you suggesting to implement the actual loop in C++ and call C# code on each iteration? Corossing native/.NET border (such as calling a .NET subrutine from C++ or vice versa) is an "expensive" operation in itself and will likely make the code slower if anything. If you are suggesting to...

Try to tack down when the order was entered into teh system and when it was executed. This will give you a better idea of what has actually happened. Was it a stop or limit order that happened to be left with the broker but you couldn't see? Or was it an order that was entered after you...

Hi Tom I haven't used "smilies" often enough to know :p means "sarcasm". I guess this is one more topic for me to catch up with... Given I have seen similar sentiment regarding the assemply language from other posters in this thread, I thought providing somewhat comprehensive a comment...

This is a common misconception. To write a fast code in assembly language requires deep understanding of how modern processors work. Also if you make suboptimal choice at any point, C++ compiler will ususally fix it for you. In assembly language you will be just stuck with a slower code...

Matlab and R are not trading strategy development platforms per se. Although with some skill you can backstest a certain subset of trading strategies with huge efficiency, these require concious effort to avoid forward-looking bias. Have you looked at TradeStation, AmiBroker, NeoTicker...

Your link included a colon at the end. Here is a correct link http://www.cnbc.com/id/41290094

Whether C# is slower or not, there is no way it's 40x300=12,000 times slower. It must be the platform doing lots of background processing (like creating "order" objects), which you don't need in a backtester written with a specific strategy in mind.

You can, and it is a better way to do it on a thinly traded instrument. However, you need good tick data for that. Not only it's more difficult to get hold of, a few data providers only provide best bid/best ask snapshot together with a trade tick, not for every change in best bid/best ask...

For a "proper" way to deal with missing data search google for "correlation with missing data" (without qoutes). The first link is a good algorithm: http://www.sitmo.com/doc/Calculating_Correlation_and_Means_with_Missing_Data

Missing minutes usually mean that there were no trades during a particular minute. So, in a way it's not "missing", it's just a way to show correctly what the market action was. Do a simpe test. For each missing minute calculate the difference between previos minute "Close" and following...

I can't say better than Elitist Trader who posted a few minutes faster than me: ... and yes, it would be $850 mln position (in gold spread, not in gold).

You can fill missing minutes with interpolation... but this will skew the results. If missing minutes are few, a better solution is to exclude all data points where one of the data points is missing. Assuming you are doing correlation on returns and not actual prices, the next step is to...

Shagi. perhaps the fund held lots of calendar spreads, not just outright futures. BTW, Daniel Shak, the fund manager in the center of the story, is a big-time "semi pro" poker player. Just a few days ago he won $1.08m in Aussie Millions Poker tournament.

Interesting article. Thanks for posting it!

To start with, 64-bit architecture lifts the 2GB limit on the amount of RAM a single application can use. With many modern computers capable of having 24GB of RAM, 2GB is really limiting. With larger amount of memory an application may be able to do in RAM things that with 32-bit architecture...

Linux vs Windows... Any difference in perforemance is gonna be minor. In both operating systems there are tons of unnecessary services. So, you might want to disable those. Desktop versions of Window are not suitable for running continuously for weeks. You need to reboot the PC regularly to keep...

If it's the best price per gugabyte then you can just get 2 or 3 of these and configure them to work as a single drive if the motherboard allows it.