Monte Carlo question

P

peppermint_tea

Hi,

I’ve been learning and researching about Monte Carlo simulations. Currently building my own one in spreadsheet form.

However, If my back test strategy was done using a risk of 1% per trade, will this have problems when trying to do Monte Carlo simulations?

I only wonder because In the back test, say I start with $10,000, risk 1% per trade and do 100 trades. By trade 90 say my equity is at $30,000 so I’m risking $300 per trade and I lose a few trades here and there.

Now when I come to do the simulator, it’s possible some of those $300 losses (which were towards the end of my back test) could come out at the start of the simulated runs. So I could get a simulated run with large losses early on but that wouldn’t happen in my real trading as I wouldn’t be risking that much early on.
Hope I have explained that well.

Just wondered what is the best thing to do in this scenario? I did think I could run the back test using a fixed number of contracts or fixed dollar amount, but I don’t trade that way, and I thought it would make sense to keep it the same as my real trading systems which are a fixed percent risk per trade.

Thanks for any suggestions or ideas
 
peppermint_tea said:
Hi,

I’ve been learning and researching about Monte Carlo simulations. Currently building my own one in spreadsheet form.

However, If my back test strategy was done using a risk of 1% per trade, will this have problems when trying to do Monte Carlo simulations?

I only wonder because In the back test, say I start with $10,000, risk 1% per trade and do 100 trades. By trade 90 say my equity is at $30,000 so I’m risking $300 per trade and I lose a few trades here and there.

Now when I come to do the simulator, it’s possible some of those $300 losses (which were towards the end of my back test) could come out at the start of the simulated runs. So I could get a simulated run with large losses early on but that wouldn’t happen in my real trading as I wouldn’t be risking that much early on.
Hope I have explained that well.

Just wondered what is the best thing to do in this scenario? I did think I could run the back test using a fixed number of contracts or fixed dollar amount, but I don’t trade that way, and I thought it would make sense to keep it the same as my real trading systems which are a fixed percent risk per trade.

Thanks for any suggestions or ideas
More...
you wrote "say I start with $10,000, risk 1% per trade and do 100 trades" and question "it’s possible some of those $300 losses (which were towards the end of my back test) could come out at the start of the simulated runs"

doesn't make sense
 
peppermint_tea said:
Hi,

I’ve been learning and researching about Monte Carlo simulations. Currently building my own one in spreadsheet form.

However, If my back test strategy was done using a risk of 1% per trade, will this have problems when trying to do Monte Carlo simulations?

I only wonder because In the back test, say I start with $10,000, risk 1% per trade and do 100 trades. By trade 90 say my equity is at $30,000 so I’m risking $300 per trade and I lose a few trades here and there.

Now when I come to do the simulator, it’s possible some of those $300 losses (which were towards the end of my back test) could come out at the start of the simulated runs. So I could get a simulated run with large losses early on but that wouldn’t happen in my real trading as I wouldn’t be risking that much early on.
Hope I have explained that well.

Just wondered what is the best thing to do in this scenario? I did think I could run the back test using a fixed number of contracts or fixed dollar amount, but I don’t trade that way, and I thought it would make sense to keep it the same as my real trading systems which are a fixed percent risk per trade.

Thanks for any suggestions or ideas
More...


In the first row you start with $10,000 so losing 1% means you lose $100, and on the second row your equity should now be $9,800, then you lose 1%, thats a -$98 loss...
 
Do you know any programming lanugages, eg python.

Just ask Chat gpt to write you a montecarlo simulator in your favourite language.

Then run it inside an online ide.

eg:

"write me a monte carlo simulator in python for a trading sytems that risk 1% and has 40% change of winning 2R and 60% chance of losing 1R"
 
%%
Like the % risk myself;
but strange how drawdowns + losses tend be worse in real trading-investing than all the back tests, forwards tests+ educated guesses. Even worse with derivatives.
Reading Rosy repost, of course the $ 300 loss or more \
could come early+ not late:caution:
I ran out of shotgun shells once on a hot game bird shoot+ doing tat all my life;
blessed with buying some more , but cant count on that.
DONT run out of bullets; UK fund manger warned ; then the elephnat comes by:caution::caution:
 
Businessman said:
You need a bigger back test. 3000+ trades, then you will have some very big drawdowns in the back test. If your back test is only 30 or 300 trades, you unlikely to have a big drawdown in your back test.
More...
%%
EXACTLY/
on that more = better.
I would also get something more than a US dollar amount, great currency, but in1928- 1933 12 eggs cost about $00.29.
ANY trade , could possibly lose it all even if rare.
So use all the data/1776 thru now /US stock market has some great trends.
I like a metals business , hit rate profit is much better than even SPY+ QQQ benchmark; but cant really scale it like can scale up SPY + such......
So , like the UK fund manager said\ dont run out of bullets/ then the elephant walks by.:caution::caution:
79or 80 year young woman got killed, in Africa by a rouge elephant this week, real time .
 
murray t turtle said:
%%
Like the % risk myself;
but strange how drawdowns + losses tend be worse in real trading-investing than all the back tests, forwards tests+ educated guesses. Even worse with derivatives.
Reading Rosy repost, of course the $ 300 loss or more \
could come early+ not late:caution:
I ran out of shotgun shells once on a hot game bird shoot+ doing tat all my life;
blessed with buying some more , but cant count on that.
DONT run out of bullets; UK fund manger warned ; then the elephnat comes by:caution::caution:
More...
%%
I had a '72 Monte Carlo back in the day, I think I paid $800 for it. Should have kept it. Classics now. Hard to back-test which cars will be classics. Buy one of these and hide it in your barn. https://en.wikipedia.org/wiki/Toyota_FJ_Cruiser.
%% Don't fill with ethanol:caution::caution:
 
vanzandt said:
%%
I had a '72 Monte Carlo back in the day, I think I paid $800 for it. Should have kept it. Classics now. Hard to back-test which cars will be classics. Buy one of these and hide it in your barn. https://en.wikipedia.org/wiki/Toyota_FJ_Cruiser.
%% Don't fill with ethanol:caution::caution:
More...
%%
LOL /I like that classic car metal > better than BA.
Even though Peter Lynch liked BA+ maybe partly because of his huge Flying Tiger winner .
Its a bit more complex than head + shoulders pattern;
i liked that watchdog pattern , for the tire dealer , her 2 ears were on full alert as she sang out a strong warning bark LOL:D:D
 
murray t turtle said:
%%
LOL /I like that classic car metal > better than BA.
Even though Peter Lynch liked BA+ maybe partly because of his huge Flying Tiger winner .
Its a bit more complex than head + shoulders pattern;
i liked that watchdog pattern , for the tire dealer , her 2 ears were on full alert as she sang out a strong warning bark LOL:D:D
More...
%%
I could never figure out why Explorers rolled over when one of those Firestone tires blew out. But that's why I'm not a lawyer I guess. Cramer has a time window on how long it takes people to forget bad things... I forget what it is(o_O).... but he used it on Chipotles, the E-coli scare..... I think Firestone took longer.
One man wanted to sell me a muffler shop a few years back.... heck of a biz, they print cash... I was dumb... I thought we'd be all electric by '25. I figured oil changes and putting plugs in Stoney's slow leak tires wouldn't cover the property taxes, let alone the help.
Canslim at IBD gives Midas a bad grade. H&R Block is investing in AI... the golden touch. I foreclosed anyway. I hope Cramer was right... don't need restless nights. :caution::caution::caution:
 
Last edited:
You must log in or register to reply here.
Back
Top