Algorithmic Trading with Python and Backtrader (Part 2)

[@parttimelarry]

Like the video? Support my content by signing up for Interactive Brokers using the link below: www.interactivebrokers.com/mkt/?src=ptlPY1&url=%2Fen%2Findex.php%3Ff%3D1338

[@iam.introspect8456]

Dude these videos are gold. So well explained and im learning a ton by following along. I did want to point out a small bug in the strategy example from the Backtrader quick-start guide. When the strategy initially checks the first days close, and then the previous two days, for the first three iterations its actually making reference to the last two days of year 2000 and comparing them to the the first day of year 2000. So to fix that i added one more if statement before checking the close values: if (len(self) >= 3): This will make sure that the strategy starts on the third trading day and makes reference to the first two days of the year, rather than the last days of the year!

[@cjhnsn20]

These are such excellent tutorials. Thank you!!

[@DeepFrydTurd]

Larry again my man your awesome !!

[@hanaanrosenthal]

Larry, you are a pleasure to learn from - thank you

[@PianoMan333]

Great series, please keep it up and go more in depth!

[@LilleyAdam]

Hi PT Larry. In your code, you used the backtrader data feed, even though you have the dataset in your current directory. Did you still need to create the class using the backtrader.feeds? Isn't that what the program uses to find the dataset?

[@henryvanvuuren2792]

Hi Larry. Run into a wall here. I imported a csv file downloaded from binance, with Generic csv. But after added strategies, nothing changed on Broker side. Pls can you help on what could be mising. It show the but and sell, lines, but cash values doesnt change accordingly.

[@LilleyAdam]

what extension have you installed that shows up all the available objects and classes?

[@s.r2350]

great vids. keep it up

[@andrewwilson4167]

Hi all. at 4 minutes I received invalid syntax error on cerebro.adddata(data). Any help would be appreciated. Thanks

[@Leo-cw6dr]

Nice tutorials! Thank you!

[@twtw785]

I got lost from the strategy part. How did you connect the overall code to the strategy (separate file)? thanks

[@oliverpolden]

I know you didn't write that TestStrategy code but isn't dataclose[0] the first element and not the current so that example is wrong?

[@LilleyAdam]

[0] is the current element according to the backtrader documents. [-1] is the first historical element.

[@oliverpolden]

@@LilleyAdam thank you for your reply. Yes you’re right about the documentation. I guess I need to do some experimenting to verify that behaviour because although it says it follows Python conventions for iterables, I can’t actually find anything in Python that says this is convention. That said, I am fairly new to Python so I’m not familiar with these conventions.

[@elcoquirivera4503]

I’m trying to wrap my head around why I would use this library. Why not just build an algorithm from scratch using a notebook ?

[@parttimelarry]

Made a whole video on frameworks recently. A couple benefits are that it can live trade your brokerage account eg. IBKR using the same Strategy class used for backtesting. Going from a notebook to actually managing Interactive Brokers connections, account balances, live order management etc takes quite some time. Also many people enjoy open source projects since there are edge cases and bugs that only surface when thousands of people have traded many different scenarios in a live environment. Also following shared design patterns is good since you can speak a common language with other developers. You can definitely do anything from scratch, but some people appreciate the the structure.

[@elcoquirivera4503]

@@parttimelarryI just watched the video and it makes sense. I was thinking of building an algorithm that would pick the optimal strategy but then I would have to pass the values back to back trader. I just decided to pursue this project so I’m working on the flow.

[@cryptopinas6616]

You are a legend.. Thanks.

[@alexaddison1783]

If this channel wasn't here there would literally be nothing on youtube for backtrader (nothing good anyways)

[@SamarthPusalkar]

Can we test intraday strategies with backtrader?

[@meetshah4800]

What if I need Indian Indices(Bank nifty) 1min time frame Backtesting???

[@meetshah4800]

Default parameters can be changed??

[@florianhillen7404]

Very nice. Test bevore you risk your hard earned money!

[@polanamis]

Great video! What IDE are you using by the way, and how are you getting those suggestions for your backtrader code as you type it?

[@hirengandhi4026]

I am guessing thats Visual Studio Code

[@Eigus-BikeCo]

Kite is a nice coding assistant for that

[@codemedio6418]

is this teach us step by step coding?

[@phaneendhraajaythota1025]

I get this below error when I try to feed data using pandas.. could you help me with this? Traceback (most recent call last): File "pandas_data.py", line 59, in cerebro.plot(style='bar') File "/home/ajay/.local/lib/python3.6/site-packages/backtrader/cerebro.py", line 996, in plot plotter.show() File "/home/ajay/.local/lib/python3.6/site-packages/backtrader/plot/plot.py", line 814, in show self.mpyplot.show() AttributeError: 'Plot_OldSync' object has no attribute 'mpyplot'

[@KundanKumar-nv6yu]

Can we backtest on 15 mins data as well for intraday strategies?

[@NoWayThisIsReaI]

The Yahoo feed only has Daily data, you need to get your own data or integrate with a provider to do that and there are documentation examples on how to do that on the backtrader website.

[@wilhelmwilhelm1251]

I always get the following error :/ AttributeError: 'str' object has no attribute 'to_pydatetime'

[@vladedoske529]

try to download csv file directly from yahoo finance

[@ИгорьАфанасьев-ы4с]

Thank you!! Very good content! I am a beginner algorithmic trader. You help a lot. There are a few questions. Can you please answer them how will the time? 1) dt = dt or self.datas[0].datetime.date(0) How to read this code? and what is dt? 2) self.log('Close, %.2f' % self.dataclose[0]) What does it mean %.2f and % self.dataclose[0]? I would be grateful for the answer

[@anouararaamouch6603]

1) equivalen to: if dt is None: dt = self.datas[0].datetime.date(0) 2) I would say check out "string and float formatters format" on G

[@shindersamra4020]

Bravo, bravo .

[@hanaanrosenthal]

Great stuff thank u

[@mank511]

Hello sir I m from India and new to coding but ur videos helping me a lot plz upload a video for login using python and also auto buy and sell using ZERODHA platform an Indian broker having API and also plz share the code ..thank u plz help