Hi all. I'm brand new to Wealth Lab. After reading his books I was led here by searching for ways to implement Laurens Bensdorp's ideas on a algorithmic/testing/trading system that also has option trading (his QuantumX system does not) because IRAs will not allow short selling, my thought being to test replacing short selling with directional options. I found this thread https://www.wealth-lab.com/Discussion/Laurens-Bensdorp-s-Automated-Stock-Trading-Systems-9377 so I am encouraged this is possible here. However, I can find no help manuals or videos on the web page to help me get started on using WL so looking for links (I must be looking right past it, sorry), any tutorials/suggestions/advice/suites to get started on implementing his system within WL, and any pro/con thoughts from users that have used it for a while here. Thanks in advance for any help!
Rename
Welcome to WealthLab! You're definitely in the right place for this kind of system development. For getting started, I'd recommend the 15-minute Quick Start video right on the WL9 home page, and the WealthLab University tutorials on our YouTube channel. There's also extensive built-in Help, and WL9 includes an AI Help Assistant that can answer questions as you're learning the program.
For the Bensdorp systems specifically, the thread you found is a great starting point. Several of the systems from his book have already been implemented as WealthLab strategies, so I'd start by downloading and experimenting with those rather than building everything from scratch.
Your idea of replacing the short side with directional options for an IRA is certainly something you could explore in WealthLab. We now have good options data support, but I'd consider that a new variation of the strategy and backtest it accordingly since options will behave quite differently from simply shorting the underlying stock.
And don't hesitate to ask questions here as you go. It's usually best to tackle one strategy or issue at a time, and there are plenty of experienced WealthLab users on the forum who can help.
For the Bensdorp systems specifically, the thread you found is a great starting point. Several of the systems from his book have already been implemented as WealthLab strategies, so I'd start by downloading and experimenting with those rather than building everything from scratch.
Your idea of replacing the short side with directional options for an IRA is certainly something you could explore in WealthLab. We now have good options data support, but I'd consider that a new variation of the strategy and backtest it accordingly since options will behave quite differently from simply shorting the underlying stock.
And don't hesitate to ask questions here as you go. It's usually best to tackle one strategy or issue at a time, and there are plenty of experienced WealthLab users on the forum who can help.
Normally, you would want to start with the most up to date version of WealthLab, which would be WL9. But if you are specifically interested in Bensdorp's system, which depends on the finantic.Indicators extension, then perhaps WL8 may be better for now. That extension is not currently available for WL9, but it will be eventually.
If I were you, I would start with WL9 anyway just to play with it. Once the finantic.Indicators extension is released for WL9, then you can start running Bensdorp's system. Both WL8 and WL9 can be installed together; it's an option.
If I were you, I would start with WL9 anyway just to play with it. Once the finantic.Indicators extension is released for WL9, then you can start running Bensdorp's system. Both WL8 and WL9 can be installed together; it's an option.
Thanks for the useful and encouraging info! I haven't yet downloaded either application. I didn't realize there was a youtube channel but I'll definitely check that out next. I guess I'm still missing something obvious, but where is the WL9 homepage you mentioned?
QUOTE:
where is the WL9 homepage
I'm not sure there's a WL9 homepage, but there is a blog article about it.
https://www.wealth-lab.com/blog/whats-new-in-wealthlab-9
WealthLab has a YouTube channel, which has an introduction to what's new in WL9.
https://www.youtube.com/channel/UCtvXnc-aiF1Gat6IthKeIjw
Honestly, I would start with WL9 and play with it (There's plenty of example strategies to play with.) until the finantic.Indicators extension gets ported to WL9. At that point, you can run Bensdorp's system. That would make the best use of your time.
The WL9 home page is the window that shows up when you open WL9.
... but if you've changed it in the Default Workspace, then click Tools > Home Page.
I've now watched all (that I can find) videos from both WL9 and Wealth Lab Strategy Training series and have learned a lot -- thanks for the pointers. I'm looking forward to learning more.
A few questions came up while watching these:
- In the training series, I thought I heard the consensus to be that short strategies do not work in the real work and that long term trend following do not either. Have I been misled by the Bensdorp books?
- I'm confused about whether a central repository site exists for strategies and portfolios of strategies. So far I have found 'published' strategies on the website (what does this mean -- published by whom? Is this meant to be exhaustive?), the presence of a patreon site, a forum post with Bensdorp strategies, and a personal blog with a 1% per week strategy. What recommendations are there for a new user to find existing strategies so as not to reinvent the wheel? Also, I'm mostly interested in suites of strategies -- is there a repository for such suites?
- And I have a philosophical question. I get that every trader has their own preferences in terms of risk, reward, comfort, complexity, time, etc., but I am confused why Bensdorps systems and philosophies (a suite of minimally correlated strategies, long/short trend following and mean reversion) are not the guiding ones in wealth lab to simply be tweaked to suit user preferences? (Or are they? From the training videos it seemed not to me.) Am I being naive that Bensdorp has a system that has proven to get 30% CAGRs in non-IRA accounts with ridiculously low drawdowns? The forum thread above indicates to me that this is the truth. Have other suites of strategies or individual strategies proven more effective? I have no attachment I'm arguing for here, I'm new to automated trading systems and read his books only this past week, so just looking to learn from the experience from others to avoid tunnel vision and dead ends that others have already found.
A few questions came up while watching these:
- In the training series, I thought I heard the consensus to be that short strategies do not work in the real work and that long term trend following do not either. Have I been misled by the Bensdorp books?
- I'm confused about whether a central repository site exists for strategies and portfolios of strategies. So far I have found 'published' strategies on the website (what does this mean -- published by whom? Is this meant to be exhaustive?), the presence of a patreon site, a forum post with Bensdorp strategies, and a personal blog with a 1% per week strategy. What recommendations are there for a new user to find existing strategies so as not to reinvent the wheel? Also, I'm mostly interested in suites of strategies -- is there a repository for such suites?
- And I have a philosophical question. I get that every trader has their own preferences in terms of risk, reward, comfort, complexity, time, etc., but I am confused why Bensdorps systems and philosophies (a suite of minimally correlated strategies, long/short trend following and mean reversion) are not the guiding ones in wealth lab to simply be tweaked to suit user preferences? (Or are they? From the training videos it seemed not to me.) Am I being naive that Bensdorp has a system that has proven to get 30% CAGRs in non-IRA accounts with ridiculously low drawdowns? The forum thread above indicates to me that this is the truth. Have other suites of strategies or individual strategies proven more effective? I have no attachment I'm arguing for here, I'm new to automated trading systems and read his books only this past week, so just looking to learn from the experience from others to avoid tunnel vision and dead ends that others have already found.
In the following, I am referring to US equities. There are short strategies that work, and there are also long-term trend-following systems that work. However, short strategies are harder to develop—if only due to the stock market's inherent long bias—and they come with different risks (theoretically unlimited risk). That being said, they can absolutely be included as an addition to a system portfolio.
You mentioned that you are still at the very beginning of your journey regarding systematic trading. I know that at the start, you don't even know where best to begin. From my perspective, it is crucial in the beginning to focus on the foundational aspects of system development, to truly understand them, and to establish a structured approach. It helps immensely if you are simply passionate and "fired up" about it, because this process will take time. Absorb as much knowledge as you possibly can. At some point, you will start to realize which information is true and which is false (you will encounter quite a few false claims).
Parallel to building your knowledge, you should test things out in WL. Don't let yourself get discouraged—the number of viable systems is small, which means you will be testing a lot of stuff just to throw it in the trash. That is completely normal. In the process, you will also discover that many common market "wisdoms" are simply incorrect (there is a reason why the vast majority of traders are unsuccessful in the long run). A prime example of this is the unconditional use of price stops.
In the end, though, it is worth it. Once you finally understand it, all doors will open for you, and you can make a lot of money with it consistently. Until then, however, it is a long road. Take your time; it really pays off. Existing systems from others can serve as an initial guide here—nothing more, nothing less.
You mentioned that you are still at the very beginning of your journey regarding systematic trading. I know that at the start, you don't even know where best to begin. From my perspective, it is crucial in the beginning to focus on the foundational aspects of system development, to truly understand them, and to establish a structured approach. It helps immensely if you are simply passionate and "fired up" about it, because this process will take time. Absorb as much knowledge as you possibly can. At some point, you will start to realize which information is true and which is false (you will encounter quite a few false claims).
Parallel to building your knowledge, you should test things out in WL. Don't let yourself get discouraged—the number of viable systems is small, which means you will be testing a lot of stuff just to throw it in the trash. That is completely normal. In the process, you will also discover that many common market "wisdoms" are simply incorrect (there is a reason why the vast majority of traders are unsuccessful in the long run). A prime example of this is the unconditional use of price stops.
In the end, though, it is worth it. Once you finally understand it, all doors will open for you, and you can make a lot of money with it consistently. Until then, however, it is a long road. Take your time; it really pays off. Existing systems from others can serve as an initial guide here—nothing more, nothing less.
I agree 100% with everything discussed in Post #8.
Test, trash, test, test, test, succeed. It's an ongoing process. Part of this effort is integrating multiple events (and indicators) together to get sufficient conformation before pulling the trigger.
The key word is "unconditional". Stop prices need to be set from a statistical prospective because their variation is a stochastic process. For example, set the stop price so there's only a 5 to 10% probability that you will sell too soon and still leave money on the table. How do you do that with WL? That's a tricky question because you need to establish a PDF (Probability Density Function; or probability distribution, which isn't Gaussian) to determine where that 5% point is, and WL can only tell you that after all the trading is done at the last bar. (WL could introduce a 2-pass backtest to return some backtest performance metrics after the first pass, but it's not doing that today. So what's the workaround?)
QUOTE:
the number of viable systems is small, which means you will be testing a lot of stuff just to throw it in the trash.
Test, trash, test, test, test, succeed. It's an ongoing process. Part of this effort is integrating multiple events (and indicators) together to get sufficient conformation before pulling the trigger.
QUOTE:
many common market "wisdoms" are simply incorrect .... A prime example of this is the unconditional use of price stops.
The key word is "unconditional". Stop prices need to be set from a statistical prospective because their variation is a stochastic process. For example, set the stop price so there's only a 5 to 10% probability that you will sell too soon and still leave money on the table. How do you do that with WL? That's a tricky question because you need to establish a PDF (Probability Density Function; or probability distribution, which isn't Gaussian) to determine where that 5% point is, and WL can only tell you that after all the trading is done at the last bar. (WL could introduce a 2-pass backtest to return some backtest performance metrics after the first pass, but it's not doing that today. So what's the workaround?)
QUOTE:
The key word is "unconditional". Stop prices need to be set from a statistical prospective because their variation is a stochastic process. For example, set the stop price so there's only a 5 to 10% probability that you will sell too soon and still leave money on the table. How do you do that with WL? That's a tricky question because you need to establish a PDF (Probability Density Function; or probability distribution, which isn't Gaussian) to determine where that 5% point is, and WL can only tell you that after all the trading is done at the last bar. (WL could introduce a 2-pass backtest to return some backtest performance metrics after the first pass, but it's not doing that today. So what's the workaround?)
I take a very pragmatic approach to this. I test different price stops and compare them to the version without any stops. If these stops don't provide a significant advantage, I leave them out (which, as a rule, is almost always the case with my strategies).
However, that is by no means always the case. There can certainly be strategies where a price stop makes sense. I just wanted to point out that many trading wisdoms (like 'always use a price stop') are simply not true.
QUOTE:
I test different price stops and compare them to the version without any stops. If these stops don't provide a significant advantage, I leave them out ...
I tend not to use trailing stops as well. But there are many positions where I lost big money because the stock took a sudden unexpected dip, which would have been prevented with a statistically computed stop.
The real problem is the sudden-dip PDF isn't Gaussian, so computing the stop using standard deviations doesn't work. You have to employ robust statistical methods, such as ranks, instead. And using ATR variations really doesn't capture sudden-dip behavior either, so there isn't any simple solution. It's a robust statistical problem involving MFE, which doesn't become available until the last backtest bar on WL. :(
QUOTE:
I just wanted to point out that many trading wisdoms are simply not true.
I totally agree. The problem is it takes about 5 or 6 years of doing backtesting to figure out what's real from what's phony. That's why we need WL.
Your Response
Post
Edit Post
Login is required