Safe way to use leverage

[stink]

Hey Tech,

I found it mate, this is the software i use not called the same thing but this is it.

Could you please have a look, its got information on the webpage about the system test functionality etc

http://www.mdsnews.com/australia/tour/default.aspx

Cheers Stink

 

[tech/a]

Hey Tech,

I found it mate, this is the software i use not called the same thing but this is it.

Could you please have a look, its got information on the webpage about the system test functionality etc

http://www.mdsnews.com/australia/tour/default.aspx

Cheers Stink

Stink.
It seems to have rudimentary functionality.
Without having the numbers reported rather than the equity chart I cant see what it actually calculates.
If its restricted to standard Oscillator codes this would be a disadvantage.
I use price action more so than oscillators.

Anyway happy to have a look at what you post up.

It wont have Montecarlo analysis however there are somethings you can do.
Will be an interesting and worthwhile exercise and I can (time permitting) test your idea when you get one you think is viable on my software (If you let me know your inputs).Please lets not test every idea!!.

Anyway will be very good for you you will learn heaps!

 

[MichaelD]

Michael 20000 portfolio tests!
Plus over the last 4 yrs none of the Systems have traded outside of their blueprint and we have had some nasty corrections!!!

I would argue that 20,000, 200,000 or 2,000,000 tests on a selected universe proves very little about overall system performance in the long term (and I don't see any nasty corrections, just a few minor blips in a raging bull market from 2500 to 5000). Anyone long term trend trading who did *not* make lots of money in the greatest bull market in history is doing something very seriously wrong.

However, I am personally becoming very aware of how easy it is to confuse the following in a bull market;
1. Brains
2. Luck
3. Positive expectancy system

Snake, I have an open mind on how the stock market works and am still pondering. Right at the moment, my beliefs are;

1. The market is a random walk with an overall positive bias.
2. There are moments of short term non-random behaviour driven by greed and fear.

To capitalize on 1 requires a robust long term exit, a la Tech Trader/ATR exit. To capitalize on 2 requires a different exit strategy, one of which I'm currently paper trading to see if I can get a positive expectancy out of it.

 

[stink]

OK tech heres a strategy i ran through the back test tonight to give you a look at how it presents the data.

the first picture is the feilds i chose to base the backtest on, and used 10k as the total capitol.

heres the output file
Summary Report By Code for - linc





Trading System Settings



Run Date: 8/30/2006 7:38:19 PM
Check Entry Signals (Custom): OXR.AX,LHG.AX,EQN.AX,PDN.AX
Initial Equity: $10,000.00
Brokerage Fee Type: Fixed
Fee Amount: $60.00
Allow Multiple entry per code: False



Total Trading System Summary



Bars in Test: 418
Total Net Profit: $6,571.345
Open Position Gain/Loss: ($16.94)
Total Commission (Fees): $6,840.00
Return on account: 65.71%

Max System Drawdown: ($1,445.22)
Max System Drawdown Percent: -22.42%


Total Trades: 65 Total Trades (Inc. Open Positions): 68
Percent Profitable: 53.85% Percent Profitable (Inc. Open Positions): 52.94%
Profit of Winners: $16,030.86 Profit of Winners (Inc. Open Positions): $16,185.90
Loss of Losers: ($9,442.575) Loss of Losers (Inc. Open Positions): ($9,614.555)
Total Winning Bars: 283 Total Winning Bars (Inc. Open Positions): 306
Total Losing Bars: 200 Total Losing Bars (Inc. Open Positions): 205


Largest Winning Trade: $1,325.22
Avg. Winning Trade: $458.025
Largest Losing Trade: ($659.94)
Avg. Losing Trade: ($314.753)
Number of Bars in Largest Winning trade: 43
Avg. Bars for Winning Trades: 8.09
Number of Bars in Largest Losing trade: 3
Avg. Bars for Losing Trades: 6.67




Trading Summary for - OXR.AX



Open Position Gain/Loss: ($112.98)
Total Commission (Fees): $1,920.00

Max Trade Drawdown: ($1,445.22)
Max Trade Drawdown Percent: -22.42%


Total Trades: 17 Total Trades (Inc. Open Positions): 18
Percent Profitable: 15.38% Percent Profitable (Inc. Open Positions): 14.71%
Profit of Winners: $5,331.86 Profit of Winners (Inc. Open Positions): $5,331.86
Loss of Losers: ($2,057.525) Loss of Losers (Inc. Open Positions): ($2,170.505)
Total Winning Bars: 116 Total Winning Bars (Inc. Open Positions): 116
Total Losing Bars: 47 Total Losing Bars (Inc. Open Positions): 52


Largest Winning Trade: $1,325.22
Avg. Winning Trade: $533.186
Largest Losing Trade: ($369.24)
Avg. Losing Trade: ($293.932)
Number of Bars in Largest Winning trade: 43
Avg. Bars for Winning Trades: 11.60
Number of Bars in Largest Losing trade: 10
Avg. Bars for Losing Trades: 6.71



Trading Summary for - LHG.AX



Open Position Gain/Loss: $155.04
Total Commission (Fees): $1,560.00

Max Trade Drawdown: ($1,439.25)
Max Trade Drawdown Percent: -22.35%


Total Trades: 15 Total Trades (Inc. Open Positions): 16
Percent Profitable: 10.77% Percent Profitable (Inc. Open Positions): 11.76%
Profit of Winners: $4,288.55 Profit of Winners (Inc. Open Positions): $4,443.59
Loss of Losers: ($2,444.84) Loss of Losers (Inc. Open Positions): ($2,444.84)
Total Winning Bars: 106 Total Winning Bars (Inc. Open Positions): 129
Total Losing Bars: 72 Total Losing Bars (Inc. Open Positions): 72


Largest Winning Trade: $1,117.25
Avg. Winning Trade: $612.65
Largest Losing Trade: ($506.85)
Avg. Losing Trade: ($305.605)
Number of Bars in Largest Winning trade: 56
Avg. Bars for Winning Trades: 15.14
Number of Bars in Largest Losing trade: 10
Avg. Bars for Losing Trades: 9.00



Trading Summary for - EQN.AX



Open Position Gain/Loss: ($59.00)
Total Commission (Fees): $2,040.00

Max Trade Drawdown: ($965.44)
Max Trade Drawdown Percent: -16.72%


Total Trades: 21 Total Trades (Inc. Open Positions): 22
Percent Profitable: 13.85% Percent Profitable (Inc. Open Positions): 13.24%
Profit of Winners: $3,020.49 Profit of Winners (Inc. Open Positions): $3,020.49
Loss of Losers: ($3,754.59) Loss of Losers (Inc. Open Positions): ($3,813.59)
Total Winning Bars: 29 Total Winning Bars (Inc. Open Positions): 29
Total Losing Bars: 74 Total Losing Bars (Inc. Open Positions): 74


Largest Winning Trade: $637.50
Avg. Winning Trade: $335.61
Largest Losing Trade: ($659.94)
Avg. Losing Trade: ($312.883)
Number of Bars in Largest Winning trade: 1
Avg. Bars for Winning Trades: 3.22
Number of Bars in Largest Losing trade: 3
Avg. Bars for Losing Trades: 6.17



Trading Summary for - PDN.AX



Open Position Gain/Loss: $0.00
Total Commission (Fees): $1,320.00

Max Trade Drawdown: ($999.97)
Max Trade Drawdown Percent: -16.96%


Total Trades: 12 Total Trades (Inc. Open Positions): 12
Percent Profitable: 13.85% Percent Profitable (Inc. Open Positions): 13.24%
Profit of Winners: $3,389.96 Profit of Winners (Inc. Open Positions): $3,389.96
Loss of Losers: ($1,185.62) Loss of Losers (Inc. Open Positions): ($1,185.62)
Total Winning Bars: 32 Total Winning Bars (Inc. Open Positions): 32
Total Losing Bars: 7 Total Losing Bars (Inc. Open Positions): 7


Largest Winning Trade: $774.71
Avg. Winning Trade: $376.662
Largest Losing Trade: ($577.92)
Avg. Losing Trade: ($395.207)
Number of Bars in Largest Winning trade: 4
Avg. Bars for Winning Trades: 3.56
Number of Bars in Largest Losing trade: 3
Avg. Bars for Losing Trades: 2.33

 

[stink]

Sorry the screen capture should have been first, but i am sure you get the picture.

Cheers Stink

 

[stink]

Oh before you ask the time period i ran the test over was from 1/1/05 to 30/8/06

 

[tech/a]

OK.

First thing I notice and the thing that is often overlooked by the un wary is the Syyatems test doesnt deduct the Brokerage cost from the Nett profit.

Take the final summary $6571 profit and $6840 Brokerage.
I checked the nett results down lower and its not taken into account.
Unfortunately a 65% winning system is infact a loser.The actual loss is A few hundred dollars.The Broker would be happy though.

Winners take losers = $6571 profit.
Brokerage of $6840 at $60 I presume $60 each way equates to 57 trades not the 65 bought and closed.So cant follow that either.

Brokerage must be considered and many many short term systems fail when brokerage is included in the nett figures.


Im not sure how the System takes trades It appears to be each bar?
That cant be right? Possibly the terminology Im not understanding.

All in all though the results are mis leading and you need to be aware of what your getting back.

Anyway let me know what Im missing or not understanding.

For interest sake here is a Tradesim report on one of the Techtrader versions

Detailed Report
(TT Master 011)

Simulation Summary
Simulation Date: 7/06/2006
Simulation Time: 5:49:40 PM
Simulation Duration: 0.51 seconds

Trade Summary
Earliest Entry Date in the Trade Database: 7/09/1998
Latest Entry Date in the Trade Database: 16/05/2006
Earliest Exit Date in the Trade Database: 28/09/1998
Latest Exit Date in the Trade Database: 19/05/2006

Start Trade Entry Date: 7/09/1998
Stop Trade Entry Date: 16/05/2006
First Entry Date: 7/09/1998
Last Entry Date: 24/03/2006
First Exit Date: 28/09/1998
Last Exit Date: 19/05/2006

Total Trading duration: 2811 days

Profit Summary
Profit Status: PROFITABLE
Starting Capital: $100,000.00
Finishing Capital: $848,414.00
Maximum Equity/(Date): $748,414.00 (19/05/2006)
Minimum Equity/(Date): -$6,752.24 (6/01/2000)
Gross Trade Profit: $883,438.60 (883.44%)
Gross Trade Loss: -$135,024.60 (-135.02%)
Total Net Profit: $748,414.00 (748.41%)
Average Profit per Trade: $4,428.49
Profit Factor: 6.5428
Profit Index: 84.72%
Total Transaction Cost: $10,140.00
Total Slippage: $0.00
Daily Compound Interest Rate: 0.0761%
Annualized Compound Interest Rate: 32.0009%

Trade Statistics
Trades Processed: 1924
Trades Taken: 169
Partial Trades Taken: 0
Trades Rejected: 903
Winning Trades: 60 (35.50%)
Losing Trades: 109 (64.50%)
Breakeven Trades: 0 (0.00%)

Normal Exit Trades: 127 (75.15%)
Delayed Normal Exit Trades: 0 (0.00%)
Open Trades: 10 (5.92%)
Protective Stop Exit Trades: 32 (18.93%)
Time Stop Exit Trades: 0 (0.00%)
Profit Stop Exit Trades: 0 (0.00%)

Largest Winning Trade/(Date): $125,685.00 (19/05/2006)
Largest Losing Trade/(Date): -$4,660.00 (7/12/2005)
Average Winning Trade: $14,723.98
Average Losing Trade: -$1,238.76
Average Win/Average Loss: 11.8861

Trade Duration Statistics
(All Trades)
Maximum Trade Duration: 1226 (days)
Minimum Trade Duration: 2 (days)
Average Trade Duration: 166 (days)
(Winning Trades)
Maximum Trade Duration: 1226 (days)
Minimum Trade Duration: 12 (days)
Average Trade Duration: 377 (days)
(Losing Trades)
Maximum Trade Duration: 222 (days)
Minimum Trade Duration: 2 (days)
Average Trade Duration: 49 (days)

Consecutive Trade Statistics
Maximum consecutive winning trades: 9
Maximum consecutive losing trades: 10
Average consecutive winning trades: 1.76
Average consecutive losing trades: 3.21

Trade Expectation Statistics
Normalized Expectation per dollar risked: $2.2000
Maximum Reward/Risk ratio: 47.50
Minimum Reward/Risk ratio: -1.26
Average Positive Reward/Risk ratio: 7.26
Average Negative Reward/Risk ratio: -0.58

Relative Drawdown
Maximum Dollar Drawdown/(Date): $14,090.41 (16/09/2002)
Maximum Percentage Drawdown/(Date): 5.8960% (6/01/2000)

Absolute (Peak-to-Valley) Dollar Drawdown
Maximum Dollar Drawdown: $18,680.36 (6.4660%)
Capital Peak/(Date): $288,896.75 (3/07/2002)
Capital Valley/(Date): $270,216.39 (1/10/2002)

Absolute (Peak-to-Valley) Percent Drawdown
Maximum Percentage Drawdown: 7.2140% ($11,960.18)
Capital Peak/(Date): $165,780.04 (18/04/2000)
Capital Valley/(Date): $153,819.86 (26/06/2001)

 

[stink]

lol yeah i am not sure either mate,

Its the first time i have used it so i would have to speak to the developers and get them to explain how it makes that calculation on the brokerage.

Anyway thanks for having a look, i am not planning on using this so it no big deal.

Regards Stink

 

[tech/a]

Im off for now but if you come up with an Idea that you think may be "The one" let me know and I'll code it up and run it if you like.(Thats if I can code it!)

 

[Bobby]

2. There are moments of short term non-random behaviour driven by greed and fear.

. To capitalize on 2 requires a different exit strategy, one of which I'm currently paper trading to see if I can get a positive expectancy out of it.

Yes Michael thats part of it, & when this happens there are others that exasperate such.

Bob.

 

[ducati916]

I would argue that 20,000, 200,000 or 2,000,000 tests on a selected universe proves very little about overall system performance in the long term (and I don't see any nasty corrections, just a few minor blips in a raging bull market from 2500 to 5000). Anyone long term trend trading who did *not* make lots of money in the greatest bull market in history is doing something very seriously wrong.

However, I am personally becoming very aware of how easy it is to confuse the following in a bull market;
1. Brains
2. Luck
3. Positive expectancy system

For the reason that, Monte Carlo is NOT STATISTICAL...........it is DETERMINISTIC

Determinism, has a very different expectation to a Statistical expectation.
The Laws of Probability & large numbers are just not as strong.
Therefore, Determinism provides an inference, suggesting further study.
Statistics provide a *Probability* to a %confidence level & standard deviations.

jog on
d998

 

[tech/a]

I would argue that any set of statistics applied---- in the end will be deterministic.
If they werent what purpose do they serve.

Semantics Ducster.

 

[stink]

OK.

First thing I notice and the thing that is often overlooked by the un wary is the Syyatems test doesnt deduct the Brokerage cost from the Nett profit.

Take the final summary $6571 profit and $6840 Brokerage.
I checked the nett results down lower and its not taken into account.
Unfortunately a 65% winning system is infact a loser.The actual loss is A few hundred dollars.The Broker would be happy though.

Winners take losers = $6571 profit.
Brokerage of $6840 at $60 I presume $60 each way equates to 57 trades not the 65 bought and closed.So cant follow that either.

Brokerage must be considered and many many short term systems fail when brokerage is included in the nett figures.


Im not sure how the System takes trades It appears to be each bar?
That cant be right? Possibly the terminology Im not understanding.

All in all though the results are mis leading and you need to be aware of what your getting back.

Anyway let me know what Im missing or not understanding.

HI Tech,

Sorry it took so long to get back to you.

I asked the question about the brokerage calculation

Hi,

The profit figure includes the fees, so the system did indeed make that profit.

If you're calculating your figures based on the total profitable and total not profitable, then keep in mind that those figures are quoted as net too, not gross. i.e. the Fee component is already included.

For a proper comparison, run the test again without the fees component. Though note too that additional transactions may be included, as extra cash might be available to take a position, where it wasn't possible before.

I realise that my response above may be a little cryptic. Please let me know if you would like a more details explanation, or feel that I've totally missed your point.

I have asked for more information, but what do you make of it?

Regards Stink