New posts

Reply to thread

Message

Hi [USER=3006]@Sir Burr[/USER]

Another great question & its a pity you have not read all my posts. This may be one post to read..
https://www.aussiestockforums.com/posts/1014617/

Strategy Development
I've posted about the extensive testing that goes into strategy development & the ongoing re-testing fine tuning selected parameters keeping in tune with the markets. Amibroker is a big scientific calculator & it's hard (for me) to code how my strategy re-balances position sizing from a direct bank feed.

Strategy capabilities
I re-testing my strategy constantly not looking at the basic code but the ongoing parameters used in my system. It should also be noted that running your strategy on different markets around the world could results in SIGNIFICANTLY different results for the same trading period & this is not a cause for concern. Running your strategy over different markets is only a minor indication of what your strategy is capable of achieving in different environments.

Start dates are important to the mix
Depending on the starting date of the test, this can also have a big bearing on the performance outcome of the strategy, this is the very reason why we need to mix the markets, mix the index & mix the dates when backtesting, don't shortcut this vital stage in the development of your strategy.

Crystal ball
Unfortunately, no one has a crystal ball, and any strategy in the real market, for some periods (sometimes long periods), may behave differently from the backtest and it is up to you to decide if you want to continue trading it or to take it off the market definitely (or simply suspend it as you did last year, continuing to monitor it, until it starts again to perform as per your expectancy).

Take the good with the bad
With a weekly strategy you have to hold the position up to week even when the price falling rapidly. With a weekly system you have to stick to the plan even if it means a bigger loss (I never override my strategy) Keep the system as simple as possible, validate it (robust backtesting) and trade it.

Make sure you test it
Stress test your strategy, backtest the strategy, test it with in-sample and out-of-sample data, and test it over different Indexes and over different time frames, even different markets around the world as your strategy needs to work under all conditions.

# The way I trade is not rocket science, it's boring as Bat$hit.

You can listen to too many
Reading various trading threads here at ASF sounds good in theory, but the problem is that getting so many different trading opinions & trading styles becomes confusing & in the end you won't know who to believe thus putting a handbrake on your trading.

Find your comfort level
By reading various trading threads you'll find out what works and what doesn't. The hardest part is finding yourself a setup you like, one that's simple and suits your mindset.

Overseas Strategy Evaluation
This is an (unaltered copy) of "one" email in a series that I'm prepared to share about the overseas testing of my Hybrid Strategy. I'm not prepared to discuss the results any further. The email returned to me is an unaltered summary of the Hybrid Strategy tested overseas on a variety of markets around the world. Conducting & believing your own research is typically fraught with danger & when my money is on the line I try not to leave anything to chance.

-----------------------------------------------------------------------------------------------------------------------------------------
email START
I found that in general, the strategy works well also in other international markets where there are a lot of inexpensive stocks. I tried some alternatives for the PositionScore but found that your criteria to be a very good fit for this system.

Thanks for the very detailed answer. Now I better understand your code and logic.

Your coding skill, in my opinion, is already pretty good since you implemented some features that are for sure far above the basics. While the original code could be refactored to achieve greater readability (something that in any case is opinionable), I did not find any major issues (the gfx part is pure aesthetics so it does not interfere with the actual trading system).
email END
-----------------------------------------------------------------------------------------------------------------------------------------

In Short - I've posted a long reply to say..
Yes I test all my strategy constantly, testing never stops.

Skate.

<blockquote data-quote="Skate" data-source="post: 1038768" data-attributes="member: 57364">Hi [USER=3006]@Sir Burr[/USER] Another great question &amp; its a pity you have not read all my posts. This may be one post to read..<a href="https://www.aussiestockforums.com/posts/1014617/" target="_blank">https://www.aussiestockforums.com/posts/1014617/</a>Strategy DevelopmentI&#039;ve posted about the extensive testing that goes into strategy development &amp; the ongoing re-testing fine tuning selected parameters keeping in tune with the markets. Amibroker is a big scientific calculator &amp; it&#039;s hard (for me) to code how my strategy re-balances position sizing from a direct bank feed.Strategy capabilitiesI re-testing my strategy constantly not looking at the basic code but the ongoing parameters used in my system. It should also be noted that running your strategy on different markets around the world could results in SIGNIFICANTLY different results for the same trading period &amp; this is not a cause for concern. Running your strategy over different markets is only a minor indication of what your strategy is capable of achieving in different environments.Start dates are important to the mixDepending on the starting date of the test, this can also have a big bearing on the performance outcome of the strategy, this is the very reason why we need to mix the markets, mix the index &amp; mix the dates when backtesting, don&#039;t shortcut this vital stage in the development of your strategy.Crystal ballUnfortunately, no one has a crystal ball, and any strategy in the real market, for some periods (sometimes long periods), may behave differently from the backtest and it is up to you to decide if you want to continue trading it or to take it off the market definitely (or simply suspend it as you did last year, continuing to monitor it, until it starts again to perform as per your expectancy).Take the good with the badWith a weekly strategy you have to hold the position up to week even when the price falling rapidly. With a weekly system you have to stick to the plan even if it means a bigger loss (I never override my strategy) Keep the system as simple as possible, validate it (robust backtesting) and trade it.Make sure you test itStress test your strategy, backtest the strategy, test it with in-sample and out-of-sample data, and test it over different Indexes and over different time frames, even different markets around the world as your strategy needs to work under all conditions.# The way I trade is not rocket science, it&#039;s boring as Bat$hit.You can listen to too manyReading various trading threads here at ASF sounds good in theory, but the problem is that getting so many different trading opinions &amp; trading styles becomes confusing &amp; in the end you won&#039;t know who to believe thus putting a handbrake on your trading.Find your comfort levelBy reading various trading threads you&#039;ll find out what works and what doesn&#039;t. The hardest part is finding yourself a setup you like, one that&#039;s simple and suits your mindset.Overseas Strategy EvaluationThis is an (unaltered copy) of &quot;one&quot; email in a series that I&#039;m prepared to share about the overseas testing of my Hybrid Strategy. I&#039;m not prepared to discuss the results any further. The email returned to me is an unaltered summary of the Hybrid Strategy tested overseas on a variety of markets around the world. Conducting &amp; believing your own research is typically fraught with danger &amp; when my money is on the line I try not to leave anything to chance.-----------------------------------------------------------------------------------------------------------------------------------------email STARTI found that in general, the strategy works well also in other international markets where there are a lot of inexpensive stocks. I tried some alternatives for the PositionScore but found that your criteria to be a very good fit for this system.Thanks for the very detailed answer. Now I better understand your code and logic.Your coding skill, in my opinion, is already pretty good since you implemented some features that are for sure far above the basics. While the original code could be refactored to achieve greater readability (something that in any case is opinionable), I did not find any major issues (the gfx part is pure aesthetics so it does not interfere with the actual trading system).email END-----------------------------------------------------------------------------------------------------------------------------------------In Short - I&#039;ve posted a long reply to say..Yes I test all my strategy constantly, testing never stops.Skate.</blockquote>

[QUOTE="Skate, post: 1038768, member: 57364"] Hi [USER=3006]@Sir Burr[/USER] Another great question & its a pity you have not read all my posts. This may be one post to read.. [URL]https://www.aussiestockforums.com/posts/1014617/[/URL] [B]Strategy Development[/B] I've posted about the extensive testing that goes into strategy development & the ongoing re-testing fine tuning selected parameters keeping in tune with the markets. Amibroker is a big scientific calculator & it's hard (for me) to code how my strategy re-balances position sizing from a direct bank feed. [B]Strategy capabilities[/B] I re-testing my strategy constantly not looking at the basic code but the ongoing parameters used in my system. It should also be noted that running your strategy on different markets around the world could results in SIGNIFICANTLY different results for the same trading period & this is not a cause for concern. Running your strategy over different markets is only a minor indication of what your strategy is capable of achieving in different environments. [B]Start dates are important to the mix[/B] Depending on the starting date of the test, this can also have a big bearing on the performance outcome of the strategy, this is the very reason why we need to mix the markets, mix the index & mix the dates when backtesting, don't shortcut this vital stage in the development of your strategy. [B]Crystal ball[/B] Unfortunately, no one has a crystal ball, and any strategy in the real market, for some periods (sometimes long periods), may behave differently from the backtest and it is up to you to decide if you want to continue trading it or to take it off the market definitely (or simply suspend it as you did last year, continuing to monitor it, until it starts again to perform as per your expectancy). [B]Take the good with the bad[/B] With a weekly strategy you have to hold the position up to week even when the price falling rapidly. With a weekly system you have to stick to the plan even if it means a bigger loss (I never override my strategy) Keep the system as simple as possible, validate it (robust backtesting) and trade it. [B]Make sure you test it[/B] Stress test your strategy, backtest the strategy, test it with in-sample and out-of-sample data, and test it over different Indexes and over different time frames, even different markets around the world as your strategy needs to work under all conditions. [B]# The way I trade is not rocket science, it's boring as Bat$hit[/B]. [B]You can listen to too many[/B] Reading various trading threads here at ASF sounds good in theory, but the problem is that getting so many different trading opinions & trading styles becomes confusing & in the end you won't know who to believe thus putting a handbrake on your trading. [B]Find your comfort level[/B] By reading various trading threads you'll find out what works and what doesn't. The hardest part is finding yourself a setup you like, one that's simple and suits your mindset. [B]Overseas Strategy Evaluation[/B] This is an (unaltered copy) of "one" email in a series that I'm prepared to share about the overseas testing of my Hybrid Strategy. I'm not prepared to discuss the results any further. The email returned to me is an unaltered summary of the Hybrid Strategy tested overseas on a variety of markets around the world. Conducting & believing your own research is typically fraught with danger & when my money is on the line I try not to leave anything to chance. ----------------------------------------------------------------------------------------------------------------------------------------- [B]email START[/B] I found that in general, the strategy works well also in other international markets where there are a lot of inexpensive stocks. I tried some alternatives for the PositionScore but found that your criteria to be a very good fit for this system. Thanks for the very detailed answer. Now I better understand your code and logic. Your coding skill, in my opinion, is already pretty good since you implemented some features that are for sure far above the basics. While the original code could be refactored to achieve greater readability (something that in any case is opinionable), I did not find any major issues (the gfx part is pure aesthetics so it does not interfere with the actual trading system). [B]email END[/B] ----------------------------------------------------------------------------------------------------------------------------------------- [B]In Short [/B]- I've posted a long reply to say.. Yes I test all my strategy constantly, testing never stops. Skate. [/QUOTE]

Verification