Hello Guys,
I am a new user of Wealth Lab.
I am testing RSI2 strategy here in Wealth Lab, however when I check the backtest results I got something different from the original Connor's technique.
The final rule for the strategy is that the indicator "close" is "greather than" indicator "SMA (5)". Well, when I check the backtest graph results, the sell order was made in the "open" of the next candle (daily time frame), not in the "close" of the same day as I wrote in the strategy rules.
There is some way to change that parameter, setting the sell order to the "close" of the same day?
Many thanks,
Daniel
I am a new user of Wealth Lab.
I am testing RSI2 strategy here in Wealth Lab, however when I check the backtest results I got something different from the original Connor's technique.
The final rule for the strategy is that the indicator "close" is "greather than" indicator "SMA (5)". Well, when I check the backtest graph results, the sell order was made in the "open" of the next candle (daily time frame), not in the "close" of the same day as I wrote in the strategy rules.
There is some way to change that parameter, setting the sell order to the "close" of the same day?
Many thanks,
Daniel
Rename
Hi Daniel,
Strategies from Blocks cannot send orders at close:
https://www.wealth-lab.com/Discussion/Buy-At-Close-in-Blocks-5650
https://www.wealth-lab.com/Discussion/Buy-Sell-at-close-and-Entry-Bar-6081
Therefore the RSI2 strategy operates with at market orders.
Strategies from Blocks cannot send orders at close:
https://www.wealth-lab.com/Discussion/Buy-At-Close-in-Blocks-5650
https://www.wealth-lab.com/Discussion/Buy-Sell-at-close-and-Entry-Bar-6081
Therefore the RSI2 strategy operates with at market orders.
Click here for a discussion how to do it by peeking with a code-base strategy:
Many thanks Eugene,
I understood, however sometimes is better to buy when market closes, because somedays it opens next day in a gap. So, I would like to compare both situations.
Eugene,
I am not able to read or write in C language, so the WLA6 was very very easier for me. Therefore I would like to know if does exist some solution comes from Wealth Lab for this kind of situation? Maybe a Wealth Lab programmer team give us support for do it these and other alterations in C language?
Many thanks,
Daniel
I understood, however sometimes is better to buy when market closes, because somedays it opens next day in a gap. So, I would like to compare both situations.
Eugene,
I am not able to read or write in C language, so the WLA6 was very very easier for me. Therefore I would like to know if does exist some solution comes from Wealth Lab for this kind of situation? Maybe a Wealth Lab programmer team give us support for do it these and other alterations in C language?
Many thanks,
Daniel
Hi Daniel,
Due to the risks involved with placing at close orders, we don't currently support it in building block strategies, but it absolutely can work in code-based. We don't want to mislead new users by thinking that simply changing the order type to "at close" and having the signals exit at close on the same bar for which you're examining the data is a simple and easy solution. It might be the easy thing for us to do, but we have more integrity than just doing the easy thing.
Consider trading such a system. When would you place your orders? You'd have to scramble a few minutes before the close of the market right? You'd then need to get the indicator values on the daily bars before the end of day data is even completed. That's currently not even supported in WL7 except for a few indicators in streaming charts. OK, so if you're able to overcome that hurdle you go ahead and place your order in time based on the incomplete indicator values. What happens if the market then turns around in the last few minutes of trading and the indicators change? You're stuck with a wrong position. Now, imagine compounding this when you want to trade 30 or 100 stocks with such a system. It's just not feasible, at least not with the current state of things.
So, we do provide the capability for more advanced users who understand the risks. They can implement a code based strategy that places "at close" order types for the same bar of data for which you're examining indicators. But, because of all the reasons I mentioned above, we don't want to make this a simple thing that any user could do and then get led down the wrong path.
I'll add - you should look at the backtest results of any strategy that is presented in this way, using "at close" trades, with a high level of scrutiny!!
Due to the risks involved with placing at close orders, we don't currently support it in building block strategies, but it absolutely can work in code-based. We don't want to mislead new users by thinking that simply changing the order type to "at close" and having the signals exit at close on the same bar for which you're examining the data is a simple and easy solution. It might be the easy thing for us to do, but we have more integrity than just doing the easy thing.
Consider trading such a system. When would you place your orders? You'd have to scramble a few minutes before the close of the market right? You'd then need to get the indicator values on the daily bars before the end of day data is even completed. That's currently not even supported in WL7 except for a few indicators in streaming charts. OK, so if you're able to overcome that hurdle you go ahead and place your order in time based on the incomplete indicator values. What happens if the market then turns around in the last few minutes of trading and the indicators change? You're stuck with a wrong position. Now, imagine compounding this when you want to trade 30 or 100 stocks with such a system. It's just not feasible, at least not with the current state of things.
So, we do provide the capability for more advanced users who understand the risks. They can implement a code based strategy that places "at close" order types for the same bar of data for which you're examining indicators. But, because of all the reasons I mentioned above, we don't want to make this a simple thing that any user could do and then get led down the wrong path.
I'll add - you should look at the backtest results of any strategy that is presented in this way, using "at close" trades, with a high level of scrutiny!!
Hello Glitch,
Thank you very much for your kindly answer.
I understood. I'll try to check the backtests using the indicator "close" as you mentioned.
Warm Regards,
Daniel
Thank you very much for your kindly answer.
I understood. I'll try to check the backtests using the indicator "close" as you mentioned.
Warm Regards,
Daniel
Your Response
Post
Edit Post
Login is required