Share: Kalman Filter Crossovers for Crypto and ‘Smart’ RollingWindows

QuantConnect Community Discussions

QUANTCONNECT COMMUNITY

Join Our Discord Channel

Join QuantConnect's Discord server for real-time support, where a vibrant community of traders and developers awaits to help you with any of your QuantConnect needs.

Draft Discussions

Bookmarked Discussions

Share: Kalman Filter Crossovers for Crypto and ‘Smart’ RollingWindows

Last comment by - July 2022

Share: Kalman Filter Crossovers for Crypto and ‘Smart’ RollingWindows

Algorithm trades Crypto using Kalman, Laguerre and EMAs for trend entry/exit signals, with SmartRollingWindow utility. Seeking community suggestions.

Share New Research

Start New Discussion Sign up

SEARCH DISCUSSIONS

377,600 Quants.

Become a Quant

VOTE FOR UPCOMING FEATURES

Share your input and vote on our future direction.

LEAN Roadmap

JOIN OUR Community MAILING LIST

Create an account on QuantConnect for the latest community delivered to your inbox.

Share: Kalman Filter Crossovers for Crypto and ‘Smart’ RollingWindows

.ekz. | ThePrinters.io | December 2021

Vladimir has shared some great filter implementations (eg: thread 1, thread 2) that inspired me to tinker with them a bit. As can be expected, it seems that Kalman, Laguerre and EMAs show favorable results when combined for trend entry/exit signals. I’ve attached a simple algorithm that trades Crypto (or any ‘trendy’ asset), using crossovers between the three. It borrows from Vladimir’s original code but it has been refactored so the logic can more readily evolve.

In the algorithm you’ll note I’m using something called a SmartRollingWindow —this is a simple utility I whipped up to more easily check for indicator crossovers and falling/rising/flat values. I hope people find it useful.

As for the strategy itself, I’d love to hear suggestions from the community on how to make it more tradable for liquid Crypto (BTC, ETH, etc). Specifically:

How might you adjust the filters? What would you change?
What regime filter would you use? Long-term MA? ADX?
How would you manage money? Volatility-based sizing?
How might you improve the exits? Trailing stops?

 Thanks in advance.

PS:

Vladimir : I took your Kalman filter logic and turned it into an indicator. It seems to work fine, but please take a look to make sure it still performs as expected.
Tagging you, Simone Pantaleoni, as I believe you're exploring similar use of for crypto trends

Author

.ekz.

December 2021

Upvote

8 people upvoted this

Vladimir INVESTOR

December 2021

ekz,

Great codding, congratulation.

Yesterday we published "Trading Kalman Filter" - the very simple version with only one indicator. and only one parameter which was not optimized yet.

Compounding Annual Return - 237.024%Drawdown - 38.500%The ratio of the annual return to the maximum drawdown is 6.16.

Compare to yours, with 3 indicator and 3 parameterCompounding Annual Return - 281.711%Drawdown - 50.300%The ratio of the annual return to the maximum drawdown is 5.60 and less than in our simple version.

In our practice, to compare algorithms, we use the ratio of the annual return to the maximum drawdown as the first metric, because we can normalize the return to the accepted maximum drawdown by simply reducing the exposure.

We usually start creating trading algorithm with the simplest version and then adding indicators and parameters one at the time only if they improve risk metrics of our choice.

Best Regards,

Vladimir

Upvote

3 people upvoted this

Mak K INVESTOR

December 2021

Hi ekz,

Very cool algo, good job. The returns are amazing but the downside risk very heavy, why don't you sacrafice some of that return to hedge your position?

Upvote

.ekz. INVESTOR

ThePrinters.io | December 2021

Hi Vladimir! Thanks for the reply and the kind words. CAR/DD ratio is a good metric to use for sure. I factor it into my algo performance criteria as well and I agree this is not tradable as is. As you rightly pointed out, for some assets using Laguerre and EMA crossing Kalman is less performant than your original version that just looks at price crossing Kalman. The fun is only just beginning, to research even better performance across the board.

On a related note: do you know how to modify the gain of the Kalman filter in Pykalman? I'd like to attempt Kalman-vs-Kalman crossovers, with more responsive vs less responsive, but to do so I need to modify the gain and I dont know how to do so with this library.

Mak K : Thanks a lot and, agreed! What are the different ways you would consider hedging positions in such a system? Would love to hear thoughts.

Upvote

.ekz. INVESTOR

ThePrinters.io | December 2021

Here’s my updated version, with Ret/DD at 9.69

CAR 299.561%
Drawdown : 30.90%

Changes:

Noticed Vladimir’s new version uses the ‘low’ of the bar for the Kalman calculation. I've now modified the Kalman indicator so it can now take any of the selectors (open, close, low, median, average, weighted, etc). Low performs the best.
I tried a few different variants of EMAs and Laguerre filter, as well as Vladimirs version of price above or below Kalman, but price crossing above/below seems to work the best.

Note: I am already doing this but noticed others might not be: setting an explicit brokerage model (instead of the default) also impacts performance. I'm using Bitfinex, where I plan to go live, and perf is significantly better.

Upvote

2 people upvoted this

Vladimir INVESTOR

December 2021

.ekz.

--> I'd like to attempt Kalman-vs-Kalman crossovers

There are thousands of ways to apply the Kalman filter to trading algorithms, from estimating the next bar's prices and the standard deviation of prices, to generating signals based on forecast errors and dynamically calculating hedge ratios.

A couple of years ago, we created over a hundred Kalman Filters applications on Quantopian that have not yet been ported to QuantConnect.

We'll see if the one you are looking for is there.

Upvote

2 people upvoted this

Vladimir INVESTOR

December 2021

Here is slightly modified “Simone Pantaleoni btcusd kalman-mom strategy”

Upvote

3 people upvoted this

Spacetime INVESTOR

December 2021

Interesting share indeed. I have ran this with ETHUSD and the results are attached. Looks like it is giving better returns. (.ekz latest version from above)

Would be nice if we can test it out with leveraged crypto futures or crypto options (maybe use it as hedging of some sort).

Upvote

1 person upvoted this

Vladimir INVESTOR

December 2021

ekz.

-> Vladimir :

for the Kalman filter, how come you have such a long warmup period?

IE: Why 5 * Period? What is the significance of the constant '5'?

From digital signal processing point, Kalman filter is adaptive infinite response filter (IIR).

Current value of any IIR substantially depends of starting value of time series.

In our practice we usually use 5 times more data points to warn up any IIR to mature.

Upvote

1 person upvoted this

.ekz. INVESTOR

ThePrinters.io | December 2021

Great to see the discussion picking up!

spacetime: Awesome that you are trying this on other cryptos. That's the idea. I’ve also tried it on ETH and SOL, and I feel the drawdown can be mitigated more.

Vladimir : Good insights. It makes sense then, that the period will influence the response filter performance. Also, thanks for the version that factors in momentum. I will play around with it, but I admit i am wary of the introduction of multiple parameters that are related to time periods (eg:lookbacks). From what I’ve learned about Kalman, a core part of its value is the fact that it adapts effectively, with much less dependency on lookback compared to, say, moving averages.On a related note, Vladimir, your current entry condition (price is above kalman) unfortunately leads to overtrading, because even though you are already invested it will sell and re-buy. This leads to about 443 trades. Adding the condition if (not self.Portfolio.Invested ) reduces this to 81 trades with slightly better performance, and a savings on transaction fees.

Trading a Crypto portfolio:
For anyone else looking to tinker with this: something to keep in mind as a ‘guiding principle’: Ideally we should able to trade this system with a portfolio of cryptos —imagine daily scanning for whichever security meets the entry criteria, and when we have several, holding them with equal weight (or some factor-based weight). In that case, we’d want a generalized system that is NOT fit to one particular security. This is why I’m hesitant to add new time based parameters because they are likely to have different requirements (e.g. the favorable momentum period of SOL will be different from that of ETH, ADA, or BTC).

Upvote

.ekz. INVESTOR

ThePrinters.io | December 2021

Updated Exits With Dual Kalmans Crossing: To reduce drawdown, I investigated other exits (other than price crossing below the Kalman --this happens quite late), and was able to test a responsive Kalman crossing below a less-responsive Kalman. I was able to increase the responsiveness by updating the transition_covariance property (higher value makes it more responsive and follow price more closely). It seems to work alright, squeezing out a better CAR/DD for BTC (298% / 29% = 10.3 ), but not so well for ETH and SOL --also, the optimized values are quite different for different securities which, as mentioned, is not ideal…

Also, because this exit signal tends to fire early sometimes, I updated the entry signal to match vladimirs, so we jump back in the trades whenever price is above the original Kalman (previously i was only entering when price *crosses* over the Kalman.

All that said, it's still worth sharing. See the attached backtest.

Upvote

Guy Fleury INVESTOR

December 2021

A small observation.

The entry and exit method 3 is the only one being executed. Methods 1, 2, and 4 do not happen.

Method 1 and 2 happen after method 3 has triggered a trade. While method 4 is never reached nor executed.

Upvote

2 people upvoted this

.ekz. INVESTOR

ThePrinters.io | December 2021

Hi Guy Fleury , great to see you chiming in!

You are correct. I parameterized entries and exits to easily test different mechanisms. Only one entry (and only one exit) method can be used at a time.

As of my last version, here are what the methods correspond to:

Entries:

Method 1: EMA crossed above Kalman
Method 2: Laguerre crossed above Kalman
Method 3: Price crossed above Kalman
Method 4: Price is above Kalman"
Method 5: Price crossed above Kalman OR (price already above Kalman & rising & Fast Kalman > slow Kalman)

Exit

Method 1: EMA crossed under Kalman
Method 2: Laguerre crossed under Kalman
Method 3: Price crossed under Kalman
Method 4: Price is below Kalman
Method 5: Fast Kalman Fast crossed under Slow Kalman

Right now, My preferred Entry/Exit pair is Entry #3 + Exit #3, as they seem to be the most consistent across securities, and require fewest total # of parameters.

Hope this helps.

Upvote

1 person upvoted this

Guy Fleury INVESTOR

December 2021

.ekz.

On your latest version above, all entries are done on method 4 while all exits are done on method 5.

Is there a reason why most of your trades occur at 20:00:00 and some at 19:00:00?

Upvote

.ekz. INVESTOR

ThePrinters.io | December 2021

Guy Fleury Again you are correct.

In my last attached backtest, I was testing entry method 4 and exit method 5. I described this in the body of the message/post. Ultimately i still believe entry #3 and exit #3 are more suited for trading multiple securities.

Not sure why some of the trades occur at 20:00 and some at 19:00. It might have to do with the way Lean consolidates the data, perhaps. Good catch, though. It might be worth investigating.

Upvote

.ekz. INVESTOR

ThePrinters.io | December 2021

In case it's not obvious to others: i am using externalized parameters, and you can see them in the left pane of the IDE.

Upvote

1 person upvoted this

Guy Fleury INVESTOR

December 2021

.ekz.

In your strategy, you have 5 trade triggers for entries and 5 for exits. The order in which they are placed can determine which will be executed first since only the first one reached of each group can be executed at any one time, as you said. You can also force which one will be executed which is what is being done here.

Theoretically, over the same lookback period, the Kalman filter is more responsive to price movements than the EMA signal which is more responsive than the Laguerre signal which in turn is more responsive than the SMA signal. If you test for the price to cross the fast Kalman filter first, the other cross-overs will not even be tested, and, therefore totally redundant. Price precedes them all.

In Vladimir's version, method 2 (Laguerre crosses Kalman) is the more productive method. The same goes for your version. However, your method 5 does outperform them all.

In this strategy using the Kalman filter as a cross-over system you simply have a variation of the old EMA cross-over trading strategy. It becomes something like finding the best lookback period for an EMA cross-over on some past market data. This might also raise a curve-fitting flag. That lookback period might not hold or be the best going forward.

Bitcoin is, let's say, volatile. It can swing 20% in a day (it has done so a few times). Any fixed lookback period could be costly. Maybe something even more adaptive might be needed, but then again, you would be back with the EMA cross-over thing. Almost suggesting that maybe minute data should be investigated next.

The strategy has no fundamental data support. The reason is simple: none is available. There is no formula that can say this is how much Bitcoin is worth. We cannot say either that Bitcoin is over-priced or under-priced at any given time. Nor can we forecast a future value within any confidence interval. You can only rely on lagging technical indicators for trading decisions as was done in your trading script. On the other hand, we could just say it is going up until it doesn't which would put you back at the right edge of the chart.

The big question is: does the price preannounce where it is going or do we just trail behind to wherever it wants to go?

I found the CAGR to be impressive. Good work.

Upvote

2 people upvoted this

Fred Painchaud INVESTOR

December 2021

Hi .ekz.,

You could also investigate variable lookback periods based on volatility (such as ATR). When volatility is higher, lookback is lower. And vice versa. It gives you faster response when needed (in big swings which normally come with greater volatility) and better statistical significance when needed (sideways market - flatter volatility).

Such as done within Dynamic Momentum Index, for instance.

Fred

Upvote

.ekz. INVESTOR

ThePrinters.io | December 2021

Guy and Fred, thanks for the feedback!

Guy Fleury (on evaluating the system):
Love the breakdown, and appreciate the detail you put in your response. You raise a very good point, too, and I landed at the same conclusion (that Vladimir highlighted to his credit): Price is the most responsive to price (ha), and so using this as a crossover signal will beat them all.
Also, you said it correctly about the Kalman crossovers, we definitely run into the problem again, with subjective parameters. because not all securities created equal... if only there was some way to normalize the lookback/parameterization... :)

Fred Painchaud (on dynamic lookbacks):
... which brings me to your point, Fred. This sounds like a good idea! Adjusting the responsiveness based on volatility, similar to how the Kaufman adaptive moving average adapts based on market efficiency (different from volatility, but comparable). With some adjustments, we may be able to get the appropriate responsiveness for the corresponding market voaltility (or market noise), regardless of the security. If trading on the daily, perhaps we will want to look at the volatility/noise on a lower time frame --perhaps the 4 hour. I will think about this and see how best to apply the logic. If you have any examples of this in practice, or any additional guidance, please share!

Everyone (on 'Strategy' vs Testbed):
Everyone, one thing to note: the backtest i've shared isnt exactly a 'strategy' as is. It's more of a strategy testbed. This is why you see multiple interchangeable entry and exit methods in there --it's easier to test and compare them against each other (using the optimizer). Typically, once I've tested and determined the ideal entry/exit method, i remove the others. I will probably remove the non-preferred ones soon, but I may want to try one or two more things first.

Upvote

1 person upvoted this

Guy Fleury INVESTOR

December 2021

.ekz.

Here are the test results of both versions (yours and Vladimir's) when forcing the program to only execute each method in turn.

The first 4 methods gave identical results. The second method was the most productive (Laguerre crossed above/below Kalman). This is a slightly delayed crossover when compared to using price. It was like shifting a position slightly forward in time.

As can be observed, your method 5 did even better.

Upvote

1 person upvoted this

Fred Painchaud INVESTOR

December 2021

Hey .ekz.,

Glad you liked the idea. It's really as old as mathematics.

An easy way to dynamically compute periods is to take one measure of volatility and divide a chosen "mid-period", adequate for the metric at hands to “dynamize”, by it. When your measure of volatility is greater than 1 (rising), your “mid-period” decreases. And vice-versa. Like pretty much every metric on y-half-bounded series like price action and volume, your measure of volatility will in theory be able to go as down as 0 and, however, won't have, in theory again, an upper limit. So you'll need to clip the resulting ratio ("mid-period"/volatility), or else, you'll end up with potentially a division by zero (and very very high resulting ratios when volatility is close to 0) and then also potentially very very low resulting ratios when volatility gets very high (the limit calculus being 0).

So, in a nutshell:

Volatility:

Can be any of the multiple metrics already known.
Say, you can base on ATR. For instance, atr(5)/sma(atr(5)). Here a period of 5 is an example. You certainly do not want to calculate volatility on a too long period since you want it to react fast.
Or you can base on close. For instance, stdev(close, 5)/sma(stdev(close, 5)).
I mean, there are multiple multiple ways to capture volatility. It's rate of change of signals. It's everywhere in engineering/math books.

Mid-period rebase:

You look at the metric you want to dynamically parametrize, your TF, and think about a “normal”, “good”, “usual” period for it. Something that is not short neither long…
You divide it by your volatility ratio above.
You clip it both min and max. So here, you think about the shortest period you want to use. And you think about the longest period you want to use.
For instance, your mid-period could be 14. 14 daily bars is typical. If you are at the minute res, 60 could be typical. ETC
So you would divide say 14 by the ratio above, floor it and make it say 3 if lower than 3 and say 52 if higher than 52. Again, numbers are examples. 3 can be much too low as a period for the dynamic metric and 52 way too high…

That's it. Well, that's one easy easy way.

Many normalize (read, rescale between 0 and 1 here - as normalization means quite a few things) volatility, usually with artificial stochastics based on running minimums and maximums (stochastics don't apply to samples. but anyway), and then multiply by the mid-period…

Electrical engineers would yet tackle the problem differently. Etc.

Enjoy!

Fred

Upvote

Nathan Swenson INVESTOR

November 2020

I'm thinking to set up the live trading, but to use it with paper trading just to get the signals in real time. I guess it would be better if it alerts ahead of time, like at open that there will be a trade at 2 hours after open. Because this strategy has so few trades, it's easy enough to just do the trades manually. I'm thinking I'll use a small amount in futures to trade NQ and ZB pair and my TD Ameritrade self managed 401k with a more conservative setup.

Regarding the Alpha Competition, this version qualifies with 84% PSR (Alpha score). Peter, you should think about entering it. You did all the work. I'm just playing around with for consideration of using for my own accounts.

Regarding the high results above, that was without any Margin. So if you wanted to go crazy, you could add some margin to it, or simply trade non-Leverage funds but lean more on margin.

Upvote

Nathan Swenson INVESTOR

November 2020

Screen shot of the end result.

Upvote

Nathan Swenson INVESTOR

November 2020

Sorry for all the posts. Having trouble posting screen shots when working from an IPAD.

Upvote

Mateusz Pulka INVESTOR

gpw radar | November 2020

Tien Duy Vo Peter Guenther

I have spotted this bug also, the problem is here (the loop is run outside the if statement):

if not self.be_in: # Close 'In' holdings #for asset, weight in self.HLD_IN.items(): # self.SetHoldings(asset, 0) #for asset, weight in self.HLD_OUT.items(): # self.SetHoldings(asset, weight) wt[self.STKS] = 0 wt[self.TLT] = .5 wt[self.IEF] = .5 # Thomas's reducing unnecessary trades for sec, weight in wt.items(): cond1 = (self.Portfolio[sec].Quantity > 0) and (weight == 0) cond2 = (self.Portfolio[sec].Quantity == 0) and (weight > 0) if cond1 or cond2: self.SetHoldings(sec, weight)

The for loop will be fire every time even if we set self.be_in to false. The correct version is:

The same should be applied in second if statement.

Upvote

1 person upvoted this

Mateusz Pulka INVESTOR

gpw radar | November 2020

Sorry some kind of formatting problem. Generally the for loop should be move inside the if statement as Tien Duy Vo pointed out when algo starts and be_in paramter is true we will buy long and short at the same time.

Upvote

Nathan Swenson INVESTOR

November 2020

I'm sorry, I don't know Python. I don't see any difference between your code blocks. Is this mistake resulting in inflated results due to margin usage?

Edit: Didn't see your last response. I will wait for a Python coder to update it. I only know C#.

Upvote

1 person upvoted this

Mateusz Pulka INVESTOR

gpw radar | November 2020

Sorry for formating problem. The code should look like this:

https://pastebin.com/cnKm9BVY

instead of this:

https://pastebin.com/pPcGFKR0

Upvote

Nathan Swenson INVESTOR

November 2020

So the issue pointed out by Gpw radar and Tien Duy Vo only occurs if "be_in = true" at the time of start. I don't see any issues with my order when ran from 2008 to 2020. It would seem that the results for mine are correct. I was worried that the results were invalid.

Upvote

Tien Duy Vo INVESTOR

November 2020

gpw radar : Thanks for pointing out. Great job. Unfortunately, the backtesting framework on Quantconnect seems to be much slower than on Quantopian. It is quite difficult to debug that way.

There is another issue related to this part of the code:

        self.Schedule.On(
            self.DateRules.EveryDay(),
            self.TimeRules.AfterMarketOpen('SPY', 120),
            self.rebalance_when_out_of_the_market
        )

        self.Schedule.On(
            self.DateRules.WeekEnd(),
            self.TimeRules.AfterMarketOpen('SPY', 120),
            self.rebalance_when_in_the_market
        )

Here, the algo is firing the two functions rebalance_when_out_of_the_market and self.rebalance_when_in_the_market, which is responsible to get "in" and "out". However, you can see that one is started every day and the other one at weekend. I wonder, whether this is done on purpose.

Upvote

Mateusz Pulka INVESTOR

gpw radar | November 2020

You are right Nathan Swenson it only occures once at the begining when you start algo and when flag in is set to true. For example check out your orders when you set the star date at 1st January 2018. It is not a big deal but worh to fix it.

Upvote

1 person upvoted this

Mark hatlan INVESTOR

November 2020

Great work! This market timer does a really nice job.

However TQQQ does do way better than SPXL. SPXL starts on 11/03/2008. UPRO starts on 6/28/2009. TQQQ starts on 2/7/2010. When switching the start date to 2/7/2010 here are the returns:

SPXL +7,552%

UPRO +7,180%

TQQQ +19,859%

I think SPXL only looked better because of the earlier first trade date.

Upvote

Nathan Swenson INVESTOR

November 2020

Thanks Mark! Good point!

Upvote

Aalap Sharma INVESTOR

November 2020

Relativily new to all this and I would really like to know are these returns realistic in the real markets? Has anyone deployed such algos to the market and seen healthy returns. Just curious...

Upvote

Nathan Swenson INVESTOR

November 2020

Aalap, I just turned the algo on and entered TMF. We shall see soon enough. I'm going very aggressive with leveraged products. It should be very realistic as it only trades about 10 times a year with trades just 2 hours after RTH market open. Algo is entirely in treasury bonds at the moment.

Upvote

Aalap Sharma INVESTOR

November 2020

Cool! I was thinking the same :)

Upvote

Mateusz Pulka INVESTOR

gpw radar | November 2020

Hi Guys,

I have added some improvments to the code and please noticed the result. So first algo has improvment in terms of open the trade at the begning of the cycle when we start the algo. The return for 2010-2020 is: 16 772%

Upvote

Mateusz Pulka INVESTOR

gpw radar | November 2020

The second run contains the following improvement. I just want to generate a signal about in/out just right after the open market. But open position just in the same way as the original algo. What is surprising the result is 21 011%. I am a little bit surprised about this, to be honest. Any idea where this difference comes from?

Upvote

Nathan Swenson INVESTOR

November 2020

Gpw Radar, I see it takes more trades when you check signal at open. Perhaps it's due to increased volatility at open which is more likely to trigger needed deviation. The 2 hours after close is actually European market close for which there is often a ramp up in index price. I had thought about messing around with that as even 5 minutes makes a difference in that period. I know 2hrs and 5 min is significant. The timing of entry vs exit and asset type all matter. Bonds seem affected more at open, mid day, and close, while 2hr after open seems more significant for equities. Lots of opportunity for improvement I believe with timing alone.

Upvote

Mateusz Pulka INVESTOR

gpw radar | November 2020

please notice that I only generate a signal in/out 1min after the market is open and then after 120min I open the position base on the signal. I have noticed that as I used IB with the cash parameter some of the orders were not filled (lack of money) and the order was later sometimes even 5 days later.

Peter Guenther Have you tried to use futures instead of etf to generate the in/out signal?

https://finance.yahoo.com/commodities?.tsrc=fin-srch

Upvote

Nathan Swenson INVESTOR

November 2020

Gpw, perhaps the delayed order in due to transaction order. I often see that the buy order is executed before the sell order for prior holdings. This probably needs to be fixed.

Upvote

Platform

Radically Open-Source Algorithmic Trading Engine

Join Our Discord Channel

Quarterly Open-Source Trading Competition

Draft Discussions

Bookmarked Discussions

SEARCH DISCUSSIONS

377,600 Quants.

VOTE FOR UPCOMING FEATURES

JOIN OUR Community MAILING LIST

IN THIS RESEARCH

PARTICIPANTS

Discussion Awards

Actions

Join QuantConnect for Free

Platform

SIGN IN

Radically Open-Source Algorithmic Trading Engine

Join Our Discord Channel

Quarterly Open-Source Trading Competition

Draft Discussions

Bookmarked Discussions

SEARCH DISCUSSIONS

377,600 Quants.

VOTE FOR UPCOMING FEATURES

JOIN OUR Community MAILING LIST

IN THIS RESEARCH

PARTICIPANTS

Discussion Awards

SHARE RESEARCH

SHARE DISCUSSION

SHARE ARTICLE

SHARE

Actions

Join QuantConnect for Free