Dear Wealth-Lab Friends Volker, Dion, and Robert,
I would also like to thank you all for the great Webinar yesterday! Really appreciate you all spending your time with us. As a reward for everyone, I tried to publish my new foolproof trading strategy called "Biff Tannen's 1 Million Percent Strategy" to the online Strategies list; however, there was an error (I guess the strategy does require uninvented technology to make it work, so I understand why it was rejected, maybe someone can use it in the future).
I also had a list of questions from the Webinar. It is highly likely that these have already been answered elsewhere or should already be known by experienced traders, so if that is the case, just tell me to keep looking, and I'll do that.
Questions:
1. Strategies:
A. Is there a way to make the indicators "either/or" instead of "all?"
B. Some strategies do not seem to ever generate sell signals, even though it is apparent that they are selling stock and buying new ones when backtesting. This was even the case with a strategy that did not set a sell condition. Do the buy conditions influence the sell conditions?
2. Backtesting:
A. Other than survivorship bias and forward-looking variables (like Chikou-span), are there other pitfalls in backtesting? What are some other common backtesting pitfalls? (Asking this so I don't get too excited about strategies that seem too good to be true.)
B. How can I get the Random data server as an option in Backtesting? I think you all sent me a link about data during the webinar, but I managed to lose it, and I couldn't find the answer looking through the documentation and past discussions (although I acknowledge the answer is probably there).
3. Optimization:
A. Sorry for this newbie question: is there a post somewhere where I can learn how to interpret the surface graph? Online searches did not help. What do the bllue/red/yellow colors represent?
B. In Walk-Forward optimization, what is the "WFO" efficiency? The documentation describes it as a percent, but it seems to be a negative or positive number. Is a positive or negative number better? Is this a meaningful way to evaluate a strategy?
4. Trading:
A. I am scared that when I put my strategy on auto-suggest or auto-trade, it will create a bunch of buy signals, and then I will buy those stocks, and then the strategy will forget when we bought the stocks, and at what price, so it won't know when to sell the stocks or how many according to the portfolio settings and so the strategy will fail. Does the strategy remember which stocks I order, and the price, and the stock purchase or sale affect on the overall portfolio? Can it query my broker to see my trading history or current portfolio composition? If the strategy is keeping track of these things automatically, can I double-check which stocks it is actively looking for sell signals on and what it thinks my portfolio looks like?
B. What happens if I have to restart the Wealth-Lab program or my computer does a restart? Is there a way to have a strategy be persistent on the Strategy Monitor and Order Manager? Or is there a way for me to restore the strategy to its most recent point before the program shut down? It seems this would be problematic if trying to make intra-day trades, as you could lose out while restarting the program/computer.
C. Why is the data source such a prominent feature of setting up a strategy on the Strategy monitor? Does it matter which data sources we use, and why can't we select our broker as a data source from the Strategy Monitor setup screen? (This is probably answered in the link I lost.)
D. Sometimes, I get buy signals in the screener (unweighted) but not in the Strategy Monitor. What am I doing wrong?
E. I have not been able to put any strategies using fundamental data into the Strategy Monitor because they require specific info (like Insider Trading or Earnings) from databases that don't seem to load into the Strategy Monitor. How can I make the Strategy Monitor talk to those databases?
5. Website:
Sometimes when trying to backtest strategies (even ones that seem legitimate to me) on the website, the following error comes up:
Could not compile the Strategy. at WealthLab.Web.Engine.BacktestingController.RunBacktest(BacktestSettingsPayload bts) in C:\Users\lucid\source\repos\WealthLab8\WealthLab.Web.Engine\Controllers\BacktestingController.cs:line 230
End Questions
I apologize for the length and breadth of these questions (which you all have probably already answered before), and I hope I haven't worn out my welcome with this first post. Thanks again for sharing your time and expertise with us yesterday.
I would also like to thank you all for the great Webinar yesterday! Really appreciate you all spending your time with us. As a reward for everyone, I tried to publish my new foolproof trading strategy called "Biff Tannen's 1 Million Percent Strategy" to the online Strategies list; however, there was an error (I guess the strategy does require uninvented technology to make it work, so I understand why it was rejected, maybe someone can use it in the future).
I also had a list of questions from the Webinar. It is highly likely that these have already been answered elsewhere or should already be known by experienced traders, so if that is the case, just tell me to keep looking, and I'll do that.
Questions:
1. Strategies:
A. Is there a way to make the indicators "either/or" instead of "all?"
B. Some strategies do not seem to ever generate sell signals, even though it is apparent that they are selling stock and buying new ones when backtesting. This was even the case with a strategy that did not set a sell condition. Do the buy conditions influence the sell conditions?
2. Backtesting:
A. Other than survivorship bias and forward-looking variables (like Chikou-span), are there other pitfalls in backtesting? What are some other common backtesting pitfalls? (Asking this so I don't get too excited about strategies that seem too good to be true.)
B. How can I get the Random data server as an option in Backtesting? I think you all sent me a link about data during the webinar, but I managed to lose it, and I couldn't find the answer looking through the documentation and past discussions (although I acknowledge the answer is probably there).
3. Optimization:
A. Sorry for this newbie question: is there a post somewhere where I can learn how to interpret the surface graph? Online searches did not help. What do the bllue/red/yellow colors represent?
B. In Walk-Forward optimization, what is the "WFO" efficiency? The documentation describes it as a percent, but it seems to be a negative or positive number. Is a positive or negative number better? Is this a meaningful way to evaluate a strategy?
4. Trading:
A. I am scared that when I put my strategy on auto-suggest or auto-trade, it will create a bunch of buy signals, and then I will buy those stocks, and then the strategy will forget when we bought the stocks, and at what price, so it won't know when to sell the stocks or how many according to the portfolio settings and so the strategy will fail. Does the strategy remember which stocks I order, and the price, and the stock purchase or sale affect on the overall portfolio? Can it query my broker to see my trading history or current portfolio composition? If the strategy is keeping track of these things automatically, can I double-check which stocks it is actively looking for sell signals on and what it thinks my portfolio looks like?
B. What happens if I have to restart the Wealth-Lab program or my computer does a restart? Is there a way to have a strategy be persistent on the Strategy Monitor and Order Manager? Or is there a way for me to restore the strategy to its most recent point before the program shut down? It seems this would be problematic if trying to make intra-day trades, as you could lose out while restarting the program/computer.
C. Why is the data source such a prominent feature of setting up a strategy on the Strategy monitor? Does it matter which data sources we use, and why can't we select our broker as a data source from the Strategy Monitor setup screen? (This is probably answered in the link I lost.)
D. Sometimes, I get buy signals in the screener (unweighted) but not in the Strategy Monitor. What am I doing wrong?
E. I have not been able to put any strategies using fundamental data into the Strategy Monitor because they require specific info (like Insider Trading or Earnings) from databases that don't seem to load into the Strategy Monitor. How can I make the Strategy Monitor talk to those databases?
5. Website:
Sometimes when trying to backtest strategies (even ones that seem legitimate to me) on the website, the following error comes up:
Could not compile the Strategy. at WealthLab.Web.Engine.BacktestingController.RunBacktest(BacktestSettingsPayload bts) in C:\Users\lucid\source\repos\WealthLab8\WealthLab.Web.Engine\Controllers\BacktestingController.cs:line 230
End Questions
I apologize for the length and breadth of these questions (which you all have probably already answered before), and I hope I haven't worn out my welcome with this first post. Thanks again for sharing your time and expertise with us yesterday.
Rename
It would be a highly unproductive approach to try and answer all the totally unrelated questions and issues, like compiler error and interpreting surface graphs, grouped together in one post. Followup Q&A will further turn such attempt into an unreadable mess.
So please break them into separate topics. I counted at least 11 potential distinct topics. Much appreciated.
So please break them into separate topics. I counted at least 11 potential distinct topics. Much appreciated.
Will do!
Your Response
Post
Edit Post
Login is required