Dump it Here

[qldfrog]

Since we are all sharing many systems (covering all the animals in Noah’s ark), I thought I would share one of my own (very simple, yet very profitable). Back-tests have been performed over the same "statistically significant" period as all the other systems presented in this thread (by others). I know back-testing means “jack”, but here are the details and results:

View attachment 118023

Trading Period = 01/01/2020 to 07/01/2021
Initial Trading Capital = $100,000
Number of Positions = 50
Index Filter = None
Universe = Any ASX Ordinary Stock (including delisted)
Shorting = No

A single run of the strategy results in the following:

View attachment 118024
View attachment 118025

A Monte-Carlo simulation (2000 run) of the strategy results in the following:

View attachment 118026

Profit Stats
Maximum Profit: $187,526.46 (187.53%)
Average Profit: $75,675.56 (75.68%)
Minimum Profit: $9,794.98 (9.79%)
Standard Deviation: $26,124.02 (26.12%)
Probability of Profit: 100.00%
Probability of Loss: 0.00%

Percent Winning Trade Stats
Maximum percentage of winning trades: 53.44%
Average percentage of winning trades: 48.26%
Minimum percentage of winning trades: 44.03%
Standard Deviation: 1.38%

Percent Losing Trade Stats
Maximum percentage of losing trades: 55.97%
Average percentage of losing Trades: 51.74%
Minimum percentage of losing trades: 46.56%
Standard Deviation: 1.38%

Maximum Peak-to-Valley Percent Drawdown Stats
Maximum Absolute Percent Drawdown: 40.4258%
Average Absolute Percent Drawdown: 29.8859%
Minimum Absolute Percent Drawdown: 15.5345%
Standard Deviation: 3.5881%

The code (Metastock + Tradesim) for this strategy is as follows:

Entry = (Ref(ExtFml("TradeSim.Rand"),-1) <= 0.05)
{If yesterday’s randomly generated number is less than or equal to 0.05, then BUY}

Exit = (Ref(ExtFml("TradeSim.Rand"),-1) >= 0.9)
{If yesterday’s randomly generated number is greater than or equal to 0.9, then SELL}

Off coarse the results could be improved by adding an index filter.

Interesting.unless you take a piss
this means that randomly selected 50 stocks kept on average 2 weeks since 01/01/2020 would have returned on average 75%
Considering xao was just below 7000 Jan last year and just above now,hard to believe unless you do invest 2k each time regardless of portfolio worth and want to demonstrate this approach.
Maybe i am just too nice and think you want to make a point?
Added: i missed the image of the cat ?

 

[Skate]

Explanation Notes

Dump it Here

Skate.

www.aussiestockforums.com

This week "HappyCat" Exploration Analysis


Skate.

 

[DaveDaGr8]

Another way to do this would be to set tradedelays (0,0,0,0) at the top and then offset your entry/exit price using

buyprice = ref(open,1);
sellprice = ref(open,1);

Doing this i ended up getting all 3 signals you're looking for on the SIGNAL day at the correct price. This backtest is actually buying on the NEXT trade open. For VMS that's the 8/1 @.055 which matches the chart.



I don't think we can have tradedelays(1,1,1,1) in the explorer because of the equity(2) command, which is needed to translate stops into sells for the explorer to pick up. It's a rabbit hole.

Another way would be to use the looping stop and use the correct entry bar, then we CAN remove the equity(2) command.

 

[Trav.]

It's a rabbit hole.

I think that sums up trading in a nutshell.

 

[dpong]

In a for loop using i++, I handled that this way. It is important to realize (I think) that there may not always be a bar for tomorrow.

Don't walk off the edge of the universe. I think.

if (i+1 < BarCount)
SellPrice = Open[i+1]; //Use tomorrow's open
else
SellPrice = Close; //Use today's close.

 

[Skate]

My previous post has been deleted as it serves no purpose.

Skate.

 

[dpong]

@Skate do you think it is possible to create a "successful" weekly system using mean reversion? My initial reaction was that mean reversion would be successful only as daily or intraday, but now I'm wondering.

Do you (or anyone else) have an opinion on viability of a weekly mean reversion strategy?

Much thanks.

 

[Skate]

@Skate do you think it is possible to create a "successful" weekly system using mean reversion? My initial reaction was that mean reversion would be successful only as daily or intraday, but now I'm wondering.

Do you (or anyone else) have an opinion on viability of a weekly mean reversion strategy?

Much thanks.

@dpong, I've traded a "daily" Mean Reversion Strategy (MRS) but not weekly. Trading with CommSec the commission rate of $29.95 each way was a killer even though the strategy was profitable. I traded the (MRS) for around 6 months a few years ago. I haven't revisited it since.

Skate.

 

[dpong]

@dpong, I've traded a "daily" Mean Reversion Strategy (MRS) but not weekly. Trading with CommSec the commission rate of $29.95 each way was a killer even though the strategy was profitable. I traded the (MRS) for around 6 months a few years ago. I haven't revisited it since.

Skate.

Thanks. I was afraid of that. Wow those are high commissions! I trade in the US and for stocks I'm paying $0. Maybe I could trade a daily system once I retire. One can dream.

 

[Trendnomics]

Interesting.unless you take a piss
this means that randomly selected 50 stocks kept on average 2 weeks since 01/01/2020 would have returned on average 75%
Considering xao was just below 7000 Jan last year and just above now,hard to believe unless you do invest 2k each time regardless of portfolio worth and want to demonstrate this approach.
Maybe i am just too nice and think you want to make a point?
Added: i missed the image of the cat ?

In terms of back-test validity and potential future viability, the Hasard Cat Daily Strategy may just be as valid as any of the other back-tested systems in this thread - the truth lies in applying an actual statistically significant period to the back-tests.

 

[dpong]

I cut and pasted incorrectly. I have corrected the code in this quotation.

      if (i+1 < BarCount)
         SellPrice = Open[i+1];  //Use tomorrow's open
      else
         SellPrice = Close[i];   //Use today's close.

I was trying to express my concern that there are cases where doing:

buyprice = ref(open,1);

...could be dangerous because if our bar is currently the last bar, trying to grab the open at last bar+1 might be an error. Using Ref( Array, N) where N is a positive number gives me pause. I did it in my code snippet only after I made sure that I wasn't working on the last bar.

If I am mistaken in my thinking, it would be a learning moment for me if someone straightened me out.

 

[Rsthree]

View attachment 118053

Explanation Notes

Dump it Here

Skate.

www.aussiestockforums.com

This week "HappyCat" Exploration Analysis
View attachment 118054

Skate.

Did I miss this week's Zebra update or has the Zebra been locked in the stable and no more petting allowed.

 

[Skate]

In terms of back-test validity and potential future viability, the Hasard Cat Daily Strategy may just be as valid as any of the other back-tested systems in this thread - the truth lies in applying an actual statistically significant period to the back-tests.

Hasard Cat Daily Strategy
@Trendnomics your backtest results are as valid as any others that have been posted in this thread. In saying this, I personally believe backtests results shouldn't to be trusted or relied upon when actually trading the strategy live.

Live trading results
Posting actual live trading results is the true measure (IMHO).

Skate.

 

[Skate]

Did I miss this week's Zebra update or has the Zebra been locked in the stable and no more petting allowed.

@Rsthree I believe posting excessively was strangling the thread. I'm posting the signals for HappyCat as @qldfrog had a sizeable investment in the strategy.

The Zebra Strategy
I wasn't going to upload the Exploration Analysis of the Zebra Strategy until someone requested it as I didn't know if there was an interest in it.

Skate.

 

[stasisbr]

I believe @Trendnomics is referring to the fact that many many systems are showing fantastic results for this year post crash however if one was to test those same systems against a wider timeframe the results may be vastly different. Hence the use of a cheeky dartboard method in the backrests (random entry and exit signals ?)

 

[Skate]

This week "Zebra Strategy" Exploration Analysis



Skate.

 

[Trendnomics]

I believe @Trendnomics is referring to the fact that many many systems are showing fantastic results for this year post crash however if one was to test those same systems against a wider timeframe the results may be vastly different. Hence the use of a cheeky dartboard method in the backrests (random entry and exit signals ?)

 

[Skate]

View attachment 118082

Fooled by Randomness
Fooled by Randomness by Nassim Nicholas Taleb

We are constantly being fooled by randomness
"Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets" is a book by Nassim Nicholas Taleb that I have mentioned once or twice before in the "Dump it here" thread. "Fooled by Randomness" was very insightful book as-well-as a good read, a real eye-opener to say the least. Taleb's book deals with the fallibility of human knowledge & it was mentioned in the video posted by @Trendnomics. This book explains why some people are successful where others aren't so lucky. Fooled by Randomness is an investigation of luck, uncertainty, probability, human error, risk, and decision-making in a world we don’t understand.

In the long term, the randomness of the markets will shrink all ego's.

The ONE Thing by Gary Keller with Jay Papasan
The surprisingly simple truth behind extraordinary results is another fantastic read.

Skate.

 

[Skate]

An insight
Let me post three charts so others can see how each position performed (TNE, PWH, & WGN)














Trading is unpredictable
The ebb & flow of the markets makes trading interesting. At times "lemons turn to lemonade" & at other times it's not always "peaches & cream". With trend trading, we get a signal to get in & a signal to get out. The period "in-between" those signals are out of our control.

Skate.

 

[Rsthree]

@Rsthree I believe posting excessively was strangling the thread. I'm posting the signals for HappyCat as @qldfrog had a sizeable investment in the strategy.

The Zebra Strategy
I wasn't going to upload the Exploration Analysis of the Zebra Strategy until someone requested it as I didn't know if there was an interest in it.

Skate.

Thank you, I too am invested in Happy Cat and happy to see it live on.
I've also been following Zebra with great interest so thank for continuing to post both.

Perhaps these live trading posts deserve their own thread.