Back

ETF trading strategy

Bump it up to 5.5 sharpe with 40% annual return. Assuming IB transaction costs and 15 cents slippage.
Update Backtest








0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


That is very cool. As a financial newbie, I'd never heard of the Hawkes process. Thanks for sharing and showing me something new to study.
0

Hawkes process isn't related to finances, it was first created for earthquake prediction. I'd suggest for you seeing the whole world as signals, so using algorithms you can analyze social physics and same algorithms can be applied to any other data, like pricing of financial instruments.
0

0

Hi, can I clone this algo to use myself?
0

Hi Robin, the small chart above has a Clone Algorithm button. I'd recommend checking out this tutorial: Introduction to the IDE
0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


Hi Satyapravin,
0

sorry about the last post....

Satyapravin, I see that you liquidate all holdings after 60 seconds. Is there a specific reason for this? I.e. have you tried other settings?

BR,
Max.
0

@Max, I was just trying something and did not have a concrete exit strategy. 60 seconds looked right since it is a temporary correction. I will let you know if I come up with a better exit strategy.

Best regards, Pravin
0

I elaborated with the time span and I achieved higher sharpe ratio when increasing it to 90 but the annual profit % dropped a bit. Looks very promising though! Looking forward to try this algo in auto trading once that feature is implemented.
0

How did you come up with the buy/sellpressure triggers (i.e. 10) ?
Optimization trap/lookback bias is on the back of my head....
But nice work anyway :)
0

I thought about that too. Since optimization isn't available so far (as far as I know) I just tried to increase and decrease it.... but so far I get the best results with 10.
0

Hi,

Made some minor changes.

Entry:
Each entry invests 25% of available cash. As money management this might seem crazy but might be acceptable due to the win rate for this strategy. If the percentage is increased to 50% the net profit gets nuts.

Exit:
As long as the asset is going in the "right" direction, depending on if we're long or short on the market, we're staying in the market. I.e. we're not liquidating strict after 60 seconds.

Thanks Satyapravin Bezwada for this strategy! I'm still trying to learn more about, and understand, the Hawkes process to fully understand this strategy. It's been a couple of years since I studied Math at the university......

BR,
Max.
0


Hi, it looks interesting, but I noticed one thing : this strategy makes entry only when calculated pips variable becomes very high !!!. I logged all entries on only SPY and
1

Hi, it looks interesting, but I noticed one thing : this strategy makes entry only when calculated pips variable becomes very high - so when we see spike in trading prices. I logged all entries on SPY and used IQFEED to get trades. All these spikes look like outliers. I'm very unsure that it will be working in reality.
2

This how this SPY 1 sec series looks for 04/08/2014

You can see outliers where this strategy generate signals.
2

Possible scenario as usually such spikes are widened sprerads, not real prices.
0

Thanks @Ardar, if that's true I'll fix it today. We build our data from raw ticks and those are genuine ticks from the exchange but they were probably cancelled later. QuantQuote has a suspicious tick flag we can use that to eliminate outliers. I'll confirm with the raw data today.
0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


Jared, I took these ticks from IQFEED and built 1 sec series on them. But it's very similar that you have same issue.
0

Interesting, I confirmed they are in the raw ticks as well but as you say they probably can't be traded in reality.

I will update the base QCAlgorithm and add a filter on the Securities Interface to delete them and let each security filter its own data. Later if you want to use different filters or remove the default filter you can override it (similar to how the Transaction Model classes operate). Sadly this might break Pravin's beautiful strategy...
0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


@Jared when will the updated QCAlgorithm be available?

BR,
Max.
0

Update Backtest





0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


Loading...

This discussion is closed