For me it's primarily been back testing on the Wizard tool. (I'm not much further into their program). Some days the back testing capability works great, others not at all, it just sits and spins. I tried to understand if there was a variable which causes this either in my computer system, TOS system, or with some setting inside Prodigio - alas, I can't find a single action which causes or resolves this problem.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Tos - Prodigio
- Thread starter RedEyeFly
- Start date
I have been working on automated trading since 2001. It is very hard to do for real.
History:
1. Started with Tradestation, it could not backtest a basket of stocks, just 1 stock at a time.
Then they are crooks, when the lowered their commision about 6 years ago in half, they did not lower mine, overcharging me by $12,000 ( I trade a lot was one of their best customers)
When I caught them they said I should have caught them sooner, and refused to give me my money back, what total crooks. So quit using them.
2. Switched to Wealthlab before the buyout by Fid. Much better backtesting, baskkets. Got to know founders pretty good, but then Fid buyout could not use IB. Fid comms much higher.
3. So went to Rightedge, did get it working with IB, but not easy as I had large list of stocks, so with 100 stocks ready to fire it chewed up a lot of RAM in the IB app it talks with. Also got to know founders of RE. Also worked with Rene Koch on codeing, but he will dissappear for weeks on end.
So with well over 1,000 hours, likley over 3,000 hours of work into autotrading I am not doing it now due to the problems/limitations with it.
Then saw Prodigo yesterday while cjhecking out TOS.
The interface looks good, love the ability to mix diffrent time frames, it appears to overcome the #1 problem of others, which it waiting till end of bar before any action can be done (Prd appears to be able to detect and act on gap ups and down). BUT saw no optimization.
IS THERE OPTIMIZATION BUILT IN (RUN 10 TESTS ON A PARAMETER SET AT 3,4,5,6,7,8,9,10,11,12) TO SEE WHAT WORKS BEST?
CAN IT PAY ATENTION TO ANOTHER STOCK OR ETF? LIKE BUY msft WHEN 6DMA CROSSES 12 DMA AND SPY 15MIN CHART RSI IS OVER 50?
I read the mesgs here, thanks for info.
History:
1. Started with Tradestation, it could not backtest a basket of stocks, just 1 stock at a time.
Then they are crooks, when the lowered their commision about 6 years ago in half, they did not lower mine, overcharging me by $12,000 ( I trade a lot was one of their best customers)
When I caught them they said I should have caught them sooner, and refused to give me my money back, what total crooks. So quit using them.
2. Switched to Wealthlab before the buyout by Fid. Much better backtesting, baskkets. Got to know founders pretty good, but then Fid buyout could not use IB. Fid comms much higher.
3. So went to Rightedge, did get it working with IB, but not easy as I had large list of stocks, so with 100 stocks ready to fire it chewed up a lot of RAM in the IB app it talks with. Also got to know founders of RE. Also worked with Rene Koch on codeing, but he will dissappear for weeks on end.
So with well over 1,000 hours, likley over 3,000 hours of work into autotrading I am not doing it now due to the problems/limitations with it.
Then saw Prodigo yesterday while cjhecking out TOS.
The interface looks good, love the ability to mix diffrent time frames, it appears to overcome the #1 problem of others, which it waiting till end of bar before any action can be done (Prd appears to be able to detect and act on gap ups and down). BUT saw no optimization.
IS THERE OPTIMIZATION BUILT IN (RUN 10 TESTS ON A PARAMETER SET AT 3,4,5,6,7,8,9,10,11,12) TO SEE WHAT WORKS BEST?
CAN IT PAY ATENTION TO ANOTHER STOCK OR ETF? LIKE BUY msft WHEN 6DMA CROSSES 12 DMA AND SPY 15MIN CHART RSI IS OVER 50?
I read the mesgs here, thanks for info.
Folks,
I just read this thread. I've been testing out Prodigio for a few days and have seen quite a few bugs relative to backtesting. At the moment for example, Trailing Stops don't work at all. I'm fearful of wasting a lot of time here. Any updates/thoughts on how much more time they really need to make this system functional? I'm willing to work around bugs, but not by losing dough or a lot of time.
Thx.
I just read this thread. I've been testing out Prodigio for a few days and have seen quite a few bugs relative to backtesting. At the moment for example, Trailing Stops don't work at all. I'm fearful of wasting a lot of time here. Any updates/thoughts on how much more time they really need to make this system functional? I'm willing to work around bugs, but not by losing dough or a lot of time.
Thx.
Truth be told, as one of the earlier posters noted, I spent considerable time trying design systems on the 'consumer' platforms (prodigio, wealthlab, ect...) which did not require any programming knowledge.
After considerable frustration I bought a few books on C++ / C# programming and hired a professional programmer to do the actual coding.
When talking to computers, it's all about accuracy and control of your program - two things which were impossible to guarantee without actually having a hand in scientific level coding, or where all of your commands are actually someone else's scripting, and consequently severely limiting.
Anyhow, hope this little bit of prospective helps.
After considerable frustration I bought a few books on C++ / C# programming and hired a professional programmer to do the actual coding.
When talking to computers, it's all about accuracy and control of your program - two things which were impossible to guarantee without actually having a hand in scientific level coding, or where all of your commands are actually someone else's scripting, and consequently severely limiting.
Anyhow, hope this little bit of prospective helps.
Just to answer a previous posters question about TOS data, yes, it's very bad. Just pull back an historical chart of the ES on a daily chart you'll find candles for Saturdays. It's really just a double of the previous Friday's candle.
They love to add features and tools, which are great for a newbie trader that wants to play and experience things for free, but everything they have is unreliable because their quotes are so unreliable. They freeze all the time for example. How could you trust an automated system with them?
It is true that most of the time, most things work all right, but to have an automated system running my money with their quotes, no way.
They love to add features and tools, which are great for a newbie trader that wants to play and experience things for free, but everything they have is unreliable because their quotes are so unreliable. They freeze all the time for example. How could you trust an automated system with them?
It is true that most of the time, most things work all right, but to have an automated system running my money with their quotes, no way.