Nowcasted data from surveys has predictive power. Surveys from Haver etc as a starting point. Lots of nuiances on the macro data and hence not a simple time series which one can apply. Let me spend some time on ur site and come back on macro series which may be helpful for your work.
Hi Celan, I got a subscription to your site. Will look through some strategies and then post on macro inputs so that I know a bit of background on the type of work you are doing.
What are your top 3 to 5 strategies I should look at?
Wow--the timing was bizarre. I sent you the note regarding the macro before seeing that you'd signed up. Like minds...;)
RE top 3 to 5 strategies. That's hard to say. It depends on what you're looking for. I would study the instrument based chart on the most recent Mudder Report, which is the forward test. In terms of macro data, I'd be interested in applying it to longer time frame data to spot trend reversals--maybe confirm a bear trap or a Wyckoff springboard?
There's a lot to digest. Let me know if you have any questions.
It's interesting that you should bring up macro and sentiment inputs. I recently saw a video of two guys that started a quant fund and they said that their secret sauce was sentiment and macro analysis. I struggle with the best way to include this data . Can you give some examples of source data that might include this?
Let me collate some material on sentiment and macro analysis, I had a large fund and we used it for slower trading and as inputs. 2 question - Can I get started on the mini subscription and if i find it useful upgrade to the full subscription by paying the difference in the price?
It's odd. I responded to both of your questions earlier but I don't see the response here. Sorry about that. Perhaps it will turn up. Meanwhile, I'll answer the question again below.
1) The answer to your first question is: all systems are based on price. Even volume indicators are used to weight price.
2) The answer to your second question is: yes of course. It is a common request, but it must be done manually. That is, Substack will not handle this sort of thing so you'll need to contact me directly by email.
Hi Celan, I was thinking about subscribing to your service. Before I do that 2 questions: 1 Are all the systems based on technicals and price action? 2 What level of strategy description does one get per strategy as a subscriber - can you share an example - anything other than Strategy 5 available? Dev
thank you very much for making freely available two of your trading strategies.
Allow me to ask a few questions and a comment related to both Strategy 1 and Strategy 5 as you present them here.
1) I think you use quite exotic technical indicators here. I have accounts at three brokers available in Poland and none of them offers VWMA or Double Stochastics, or RSS. Of course, I can calculate them on my own but this will inevitably lead to the delay in the trading since my brokers don't offer any API for algorithmic trading. This is not a criticism, just a statement of fact which you might want to comment about.
2) Your strategies contain very specific value of parameters (like 10 or 40) that are to be hard-coded. They seem to be obtained by an optimization process. If so, aren't you afraid of overfitting?
3) Are the above parameters independent of the trading time scale? Will they work for both 5-minutes and weekly data?
4) You have performed a very remarkable work on backtesting. But have you also tried your strategies on synthetic (as opposed to historical) data?
5) What exactly do you mean by "Volume-Weighted Moving Average" (refers to Strategy 1)? I can imagine at least several different definitions of what that might be. Is it the same as Volume-Weighted Average Price?
6) I don't understand the sentence that begins with "When MACD (MACD plot) crosses above MACD (Avg plot), with a look back period of 2 (...)". Does that mean that instead of the standard signal line of MACD, a 2-period moving average of MACD is to be used?
2) No, strategies are not hard-coded. We are also very concerned about overfitting. You should read our FAQs. Here's what I said about overfitting:
What is overfitting and what are you doing about it?
Another commonly asked question is in regards to overfitting. You can read the following posts to learn more about overfitting and what we’ve done to reduce its impact: Overfitting: What is it and what can we do about it and What Are We Doing To Ensure Backtest Accuracy? A few months after writing these posts I received a question regarding strategy strength and what optimization tools we’re using to increase robustness. I answered that question in the post: How Can You Tell If An Automated Trading Strategy Will Perform Well Over Time?
I tried testing this: https://medium.com/@eloisecha/i-test-you-profit-trading-strategy-2-522d7a2590a1
Nowcasted data from surveys has predictive power. Surveys from Haver etc as a starting point. Lots of nuiances on the macro data and hence not a simple time series which one can apply. Let me spend some time on ur site and come back on macro series which may be helpful for your work.
Hi Celan, I got a subscription to your site. Will look through some strategies and then post on macro inputs so that I know a bit of background on the type of work you are doing.
What are your top 3 to 5 strategies I should look at?
Also, pls contact me directly or on a private strategy. This is a public strategy.
AutomatedTradingStrategies@protonmail.com
Hi Dev,
Wow--the timing was bizarre. I sent you the note regarding the macro before seeing that you'd signed up. Like minds...;)
RE top 3 to 5 strategies. That's hard to say. It depends on what you're looking for. I would study the instrument based chart on the most recent Mudder Report, which is the forward test. In terms of macro data, I'd be interested in applying it to longer time frame data to spot trend reversals--maybe confirm a bear trap or a Wyckoff springboard?
There's a lot to digest. Let me know if you have any questions.
Celan
Hi Dev,
1) I would say 100% of the strategies are based on price;)
2) Yes of course, but I will have to do that manually. It is not handled by Substack. Just send me an email: AutomatedTradingStrategies@protonmail.com
Thank you for posting this Celan very prompt and interesting work that you are doing.
1 The systems which you present are all based on prices and volumes as input? do they consider any other input data - macro, sentiment, etc
2 Can I get started on the mini subscription and if i find it useful upgrade to the full subscription by paying the difference in the price?
It's interesting that you should bring up macro and sentiment inputs. I recently saw a video of two guys that started a quant fund and they said that their secret sauce was sentiment and macro analysis. I struggle with the best way to include this data . Can you give some examples of source data that might include this?
Let me collate some material on sentiment and macro analysis, I had a large fund and we used it for slower trading and as inputs. 2 question - Can I get started on the mini subscription and if i find it useful upgrade to the full subscription by paying the difference in the price?
Any update on that material?
It's odd. I responded to both of your questions earlier but I don't see the response here. Sorry about that. Perhaps it will turn up. Meanwhile, I'll answer the question again below.
1) The answer to your first question is: all systems are based on price. Even volume indicators are used to weight price.
2) The answer to your second question is: yes of course. It is a common request, but it must be done manually. That is, Substack will not handle this sort of thing so you'll need to contact me directly by email.
Hi Celan, I was thinking about subscribing to your service. Before I do that 2 questions: 1 Are all the systems based on technicals and price action? 2 What level of strategy description does one get per strategy as a subscriber - can you share an example - anything other than Strategy 5 available? Dev
Hi Dev,
Take a look at #1 and I also recently unlocked #3, which I'm sending to you now.
Let me know if you have any other questions,
Celan
just curious how many subscribers do you have? approx?
can you please lower your subscription cost from 1249$ to something more reasonable, like 100$ per month???
Check out ATS Mini. It's the lower cost version of ATS: https://atsmini.substack.com/
Also, if you can get together with a group of 4 or more, I will give you an 80% discount. https://automatedtradingstrategies.substack.com/subscribe?coupon=10cb1667&group=true
Dear Mr. Bryant,
thank you very much for making freely available two of your trading strategies.
Allow me to ask a few questions and a comment related to both Strategy 1 and Strategy 5 as you present them here.
1) I think you use quite exotic technical indicators here. I have accounts at three brokers available in Poland and none of them offers VWMA or Double Stochastics, or RSS. Of course, I can calculate them on my own but this will inevitably lead to the delay in the trading since my brokers don't offer any API for algorithmic trading. This is not a criticism, just a statement of fact which you might want to comment about.
2) Your strategies contain very specific value of parameters (like 10 or 40) that are to be hard-coded. They seem to be obtained by an optimization process. If so, aren't you afraid of overfitting?
3) Are the above parameters independent of the trading time scale? Will they work for both 5-minutes and weekly data?
4) You have performed a very remarkable work on backtesting. But have you also tried your strategies on synthetic (as opposed to historical) data?
5) What exactly do you mean by "Volume-Weighted Moving Average" (refers to Strategy 1)? I can imagine at least several different definitions of what that might be. Is it the same as Volume-Weighted Average Price?
6) I don't understand the sentence that begins with "When MACD (MACD plot) crosses above MACD (Avg plot), with a look back period of 2 (...)". Does that mean that instead of the standard signal line of MACD, a 2-period moving average of MACD is to be used?
I thank you for your kind attention and remain
yours sincerely,
Maciej Janowicz
Hi Maciej,
1) Can you use Ninjatrader 8 in Poland?
2) No, strategies are not hard-coded. We are also very concerned about overfitting. You should read our FAQs. Here's what I said about overfitting:
What is overfitting and what are you doing about it?
Another commonly asked question is in regards to overfitting. You can read the following posts to learn more about overfitting and what we’ve done to reduce its impact: Overfitting: What is it and what can we do about it and What Are We Doing To Ensure Backtest Accuracy? A few months after writing these posts I received a question regarding strategy strength and what optimization tools we’re using to increase robustness. I answered that question in the post: How Can You Tell If An Automated Trading Strategy Will Perform Well Over Time?
https://automatedtradingstrategies.substack.com/p/frequently-asked-questions?s=w
3) The performance data is based on the data series listed on the performance chart. See the second column in the performance chart: https://automatedtradingstrategies.substack.com/p/strategy-descriptions
4) Why would we use synthetic data over historical data?
5) Here's an article by Ninjatrader on VWMA: https://ninjatrader.com/blog/what-is-a-volume-weighted-moving-average-vwma/
6) A look back period of 2 means the strategy will look at the last two periods to see if the command has been triggered.
Hope that helps,
Celan