This model is optimized for SPY on a daily time-frame.
Even though it is still profitable (Profit factor > 1) on other time-frames, such as 1h or weekly, I strongly advise you to NOT consider these signals.
You might also get positive returns on other assets, and time-frames, and I also strongly advise you to NOT consider them for your trades. For example: AAPL-1h GOOGL-D-W TSLA-D-W PYPL-D INTC-W MSFT-D-W FDN-D-W And so on …
This model is an optimization (parameters tuning) of a meta-model (generic model) for the SPY. It is mainly based on a conjunction of price & volume personal indicators for both entry and exit signals.
The relative portability of the model to other assets and time-frames, coupled with a "Development set -> Validation set" approach, confers it a stronger reliability, and a better warranty of not being « over-optimized ». The meta-model has also served for other model buildings, about 100 as of today.
Be advised that this model applied to real data will get much lower profit factors. During high-volatility periods (such as current times), the model might also be less accurate, as "News streams", more than "prices and volumes", make the market.
As always, this model is for an educational purpose only, and should never be considered as a single decision tool. So, study it, and make sure your decisions are still your own choice.
릴리즈 노트
⋅
Update to comply with the last data.
AGAIN PLEASE DON'T CONSIDER IT THESE DAYS. VOLATILITY IS TOO HIGH.
릴리즈 노트
⋅
Version which takes into account last 2 months of data
릴리즈 노트
⋅
New version with latest data
릴리즈 노트
⋅
New version with latest data
릴리즈 노트
⋅
1) Added stopLoss exits for both LONG and SHORT signals 2) Narrowed the spectrum of each substrategy to the best of the best 3) Added a max VIX parameter to avoid false entry/exit signals due to high volatility 4) Removed the swing-short substrategy which was not convincing enough on real data
릴리즈 노트
⋅
(Republished for the daily Timeframe, sorry)
1) Added stopLoss exits for both LONG and SHORT signals 2) Narrowed the spectrum of each substrategy to the best of the best 3) Added a max VIX parameter to avoid false entry/exit signals due to high volatility 4) Removed the swing-short substrategy which was not convincing enough on real data
After a few likes recently, I was wondering how it had performed since my last update about a year ago.
So:
9 trades, just a few days each
A profit factor of 4
66% positive trades
Just as expected. See you in 1 year.
Email-Trader
⋅
Please don't use this model, nor any other one, during high volatility periods. They are not build upon these rare scenarios. Consider the signals solely with a VIX under 28 max.
My personal guess is that we should be going toward a great depression for a few months. So, be patient.
Email-Trader
⋅
Please thumb it up if you like it!
SorosBrezekhiel
⋅
Dear Harold Is it Repainting? Or working this well in real scenario? thx
Email-Trader
⋅
@SorosBrezekhiel, No repainting ! However, a model is still a model and does not work well all the time. It may getting old at some point in the future. Signals are less reliable in case of strong volatility. Check how it performs since last update and see by yourself.
SorosBrezekhiel
⋅
@HaroldChaat, Thanks Harold, however backtesting do not work correct? when I add init capital to 1000 it show nice % but when use capital 10 000 or higher it goes to 5% only, why? Thanks
d-seven
⋅
Good stuff.
How does it perform vs B&H?
d7
dagekko
⋅
Short signal just triggered today - following to see b/c I am see a drop coming in the near future.
Email-Trader
⋅
@dagekko, Yep I know. It was triggered yesterday at closing on my own "study" pine script.
But be careful, FED is buying the deeps with "unlimited" funny money, making short positions very risky.
dagekko
⋅
@HaroldChaat, Looking great so far today. I'm paper trading this with options about a month out ATM - Would be up 20% today!
So:
9 trades, just a few days each
A profit factor of 4
66% positive trades
Just as expected. See you in 1 year.