Algorithmic Trading in Python - Bollinger Bands, Trading strategy and Backtest
HTML-код
- Опубликовано: 20 июн 2024
- In this video we will take a look at coding Bollinger Bands in Python, do some amendments and backtest a strategy based on BBs.
Get the Notebook/Source code by becoming a Tier-2 Channel member:
/ @algovibes
Video on coding a Trailing Stop Loss:
• Cryptocurrency 💰 Trad...
Be kindly invited to incorporate Bolling Band Signals to the Recommender system we have built here:
• How to build a Stock R...
If you can extract value out of my videos I would really appreciate you subscribing! Thanks a lot :-)
0:00 - 02:18 Introduction
02:18 - 07:20 Coding the calculation of Bollinger Bands
07:20 - 12:38 Plotting Bollinger Bands
12:38 - 13:19 Problem of multiple buying/selling signals
13:19 - 18:50 Coding the strategy (Crossing)
18:50 - 24:12 Backtest
24:12 - 25:00 Final thoughts
This video is not an investment advice and is for educational and entertainment purposes only.
#Python #Trading #BollingerBands
![Backtesting Rayner Teos RSI SMA Trading Strategy in Python [Part II]](http://i.ytimg.com/vi/rYfe9Bg2GcY/mqdefault.jpg)
![Backtesting Rayner Teos RSI SMA Trading Strategy in Python [Part II]](/img/tr.png)
![Backtesting Rayner Teos 1689% Bollinger Bands Trading Strategy in Python [Part II] - OVER 1K stocks](http://i.ytimg.com/vi/d4WgOIrby5Q/mqdefault.jpg)
You are amazing, today I'm doing a robo advisor, so I take some popular indicators and this video opens me a great outlook of Bollinger bands, thank you very much!
Thx mate! Appreciate you leaving a comment
Here again , thank you pal
A very comprehensive tutorial... Thank you very much. Would also love to see a beginner trade terms broken down (also in codes).
Thank you so much
Welcome buddy, thanks a lot for watching!
Your explanation is clear as water
Great to hear, thank you! :-)
You’re a legend 🙌🙌🙌
You are for supporting me (-:
Brilliant
Nice, really good, more video like that
Thanks a lot for your feedback :-)
Massive thank you :)
Thanks a lot for watching Oscar!
Thanks so much for the vids man your the man
Thank YOU for your support :-)
Thank you so much ☺️
welcome buddy. Be kindly invited to check out my other stuff :-)
Great coding sir,
She'll it work for indian indexs
Thanks buddy. Well, try it out :D
BTW if you need Indian stock market data or are interested in a momentum strategy on the Indian market I got you covered:
ruclips.net/video/YIsKSQh1xpY/видео.html
Wonderful video, I subscibed. I am a complete new beginner to python. I want to start learning, so that even i can do these programs.
How can I start of with learning. Any specific playlists to watch in your channel or any recommendation ?
Thanks for your subscription buddy, appreciate it :-)
You can check out my Python Introduction playlist and my Python for finance playlist.
More advanced stuff can be found e.g. in the stock recommender series.
Wishing you success and joy on your journey!
Good vid.... Thanks.
Thanks buddy :-)
:) I want more. I tried TradeWindow etc now testing PlutoHQ Trading
Thank you.
Thank YOU :-)
could you maybe post your code in the video description? would make it easier to copy it and tinker around along your videos! Thx mate!
Nice one. I wonder whether you could make it so that the signal triggers when the crossover actually occurred. Your current code only shows the data point that is below the BB but it doesn't imply the crossover has occurred.
Thanks mate, sure that's possible. You would just need to check if the current price is above/below the upper/lower band.
I think it is better to let the signal of BB to trigger a trailing stop and put a condition for buy or sell signal, once the trailing stop is triggered(position close) then enter the new position of signal entry... hope that make sence 😆
Video is great and in line with your other videos. Can you do one for rsi + divergence strategy as well if possible
Thank you mate. Already did here: ruclips.net/video/r8pU-8l1KPU/видео.html
Fab video! Could you make a video on how to optimise the parameters of technical indicators? For example, finding what is the most profitable window and standard deviation for bollinger bands?
Hi buddy, thanks a lot :-)
I already covered the main concepts of that in this one here:
ruclips.net/video/vWVZxiaaTCs/видео.html
But there surely will be more on that.
@@Algovibes Thanks for your answer, found this video yesterday. Keep up the good work, loved your videos! Hopefully, there will be more people interested in algos and you will gain more subscribers soon!
Thanks.
Thank YOU :-)
Interested in your background and what led you to algo trading. Brilliant video btw
Hi buddy, thank you very much! :-)
Let me try to break it down shortly:
1. I got deeper into this topic while "writing" (rather coding) my master thesis
2. I am privately into trading both stocks and cryptos since several years
3. Have a strong interest in programming and tech topics in general.
Also it is pretty nicely combinable with my current job which is in the field of Data Science.
Hope this is giving you some background!
BTW I remember the greeks (options) from an internship and your name reminds me of it :-D
@@Algovibes Haha, yeah my name was more fun back in 2020 when options were trading a higher premiums because of all the volatility, able to pull 5% a month. Now option selling strategies alone don't provide as good a returns as they did.
Thanks for your background. It's cool that you decided to get into this area. Definitely difficult, but even more rewarding. Wish you all the success.
Hi, great video.
I have been following Binace Trading in Python.
is i possible to provide some tips to have Bollinger Bands strategy using Cripto (Binance)?
I am not sure how to do it because in previous videos you are getting only prices with Websockets-
Its not clear for me how to achive this. could I use websockets or need I to change back to the method used in very first binance video?
Just I have seen anew video for crypto with SMA. Probably I will get more tips there
Hi buddy, thanks a lot!
No you can actually do that with streamed data.
Great video , could you do a video on good projects that one can work on (similar to this (algo trading)) to get a job in quant finance/ algo trading.
Thanks for watching and your suggestion. Appreciate it :-)
Will see what I can do but here are some thoughts from my side:
Quantitative finance is a competitive field where in general a technical degree (Master or even PhD) is mandatory.
Anyhow, when doing projects some general stuff:
1. Write down in words what you want to trade
2. Define an entry and exit strategy (including risk management)
3. Code the strategy
4. Backtest it
5. Actually test it using an API in the best case
6. Automate it by deploying the strategy (Cloud or similar)
🔥
:-)
Great video, as usual...However, it looks like your loop is only running if the initiating position is a buy. But what happens if the first signal is a sell? You will need to add True to the open_pos. on an opening sell signal. I guess you are only running this to look for opening buy positions, but you might lose out on valuable initiating sell signals. I think that needs to be added to your loop. Just a thought....that video. and coding is great. Keep up the great work
Thank you for your kind words and thank you even more for sharing your thoughts! :-)
You are right! I actually did that on purpose to only include "complete" trades (so no short selling etc.) but I totally agree with what you said.
@@Algovibes same problem can you send updated code? thanks
# For generate signals which is buy,sell,buy,sell....(orderly)
# we don't wanna see consecutive buy or sell signals
buys = []
sells = []
open_pos = False
for i in range(len(df)):
if df.upper[i] < df[col][i]:
if df.upper[i] < df[col][i]:
if open_pos == False:
sells.append(i)
open_pos == True
elif df.lower[i] > df[col][i]:
if open_pos:
buys.append(i)
open_pos = False
elif df.lower[i] > df[col][i]:
if df.lower[i] > df[col][i]:
if open_pos == False:
buys.append(i)
open_pos == True
elif df.upper[i] < df[col][i]:
if open_pos:
sells.append(i)
open_pos = False
sorry actual solution:
def implement_bb_strategy(data, lower_bb, upper_bb):
buy_price = []
sell_price = []
bb_signal = []
signal = 0
for i in range(len(data)):
if data[i-1] > lower_bb[i-1] and data[i] < lower_bb[i]:
if signal != 1:
buy_price.append(data[i])
sell_price.append(np.nan)
signal = 1
bb_signal.append(signal)
else:
buy_price.append(np.nan)
sell_price.append(np.nan)
bb_signal.append(0)
elif data[i-1] < upper_bb[i-1] and data[i] > upper_bb[i]:
if signal != -1:
buy_price.append(np.nan)
sell_price.append(data[i])
signal = -1
bb_signal.append(signal)
else:
buy_price.append(np.nan)
sell_price.append(np.nan)
bb_signal.append(0)
else:
buy_price.append(np.nan)
sell_price.append(np.nan)
bb_signal.append(0)
return buy_price, sell_price, bb_signal
buy_price, sell_price, bb_signal = implement_bb_strategy(df[col], df['lower'], df['upper'])
hello,when i type import yfinance or any other import it says"import yfinance could not be found (reportmissingimport), i am not sure why
This problem occurs because you are not writing code in a Python interpreter.
My guess is that you are using VS Code but I might be wrong with that.
In case I am right, check this out:
code.visualstudio.com/docs/python/environments
Ty for your videos! I've been following along with several of them and I've been learning alot! As of yesterday though the yfinance API doesn't seem to work for me :( Same for you?
Awesome. Thanks a lot for watching them and your comment :-) Please check my newest community post. I recently had problems with yfinance as well and solved it with the described step.
@@Algovibes You're the best! This solved my problem! Bonus fun fact, your voice and english dialect sounds exactly like one of my online gaming friends. Watching your videos is like taking lessons from a more tech savvy version of him haha!
nice! what IDE is this btw?
Thanks man, just Jupyter Notebook.
Hey, I just made a similar algo that uses bollinger in crypto but Im having a issue while trying to keep my client going. It seems like my stream just get dropped after some hours. How can I make it run indefinitely? Thank you for your videos as always
Hi, which API are you using?
@@Algovibes Im using the binance client, Im not sure what is his name, its the one that you put your API key and API secret. I request data every minute, which I think its pretty far from the request limit, but still after around 15 hours running I get dropped. I get the error "connection forcibly closed by remote host", but I dont think Im doing anything irregular . Thanks
nice video, but is there a way to make a video about bollinger bands %b?
Hey man, thanks a lot. Didn't plan on that yet but noted it.
Great video, unfortunately loop in not working correctly in my case. It returns only buy signals and empty sell list. Please advise
Thanks for your comment, Spasibo :-) Probably a wrong indentation. Can you post your code?
@@Algovibes
buys = []
sells = []
open_pos = False
for i in range(len(df)):
if df.Lower[i] > df.Close[i]:
if open_pos == False:
buys.append(i)
open_pos == True
elif df.Upper[i] < df.Close[i]:
if open_pos == True:
sells.append(i)
open_pos ==False
It's a bit late. But the issue is when you are setting open_pos. open_pos == true does not set the value of open_pos it returns whether or not open_pos is true. You need one equals sign to set the value. open_pos = True. You also need to fix this is lower block where you sell and set open pos false.
Hey thanks for sharing this. Much appreciated!
I am facing some Json error while downloading data from yfinance. Please help me with this
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)
Hi buddy, thanks for your comment. Appreciate it! :-)
Please check my newest community post. I had this problem as well and I got it working again.
Would be helpful if you can confirm this is solving the problem for you as well (don't forget to restart Jupyter after updating).
@@Algovibes
Thanks for the solution. Your method is worked on solving JSONDecodeError, I guess.
But I am facing a new error here. Can you please help me out on this as well?
requests.exceptions.SSLError: HTTPSConnectionPool(host='query2.finance.yahoo.com', port=443): Max retries exceeded with url: /v8/finance/chart/MSFT?period1=1609439400&period2=1627360626&interval=1d&includePrePost=False&events=div%2Csplits (Caused by SSLError(SSLCertVerificationError(1, '[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self signed certificate in certificate chain (_ssl.c:1125)')))
@@ragha2846vs that's just an internet connection error
Very good video and your progrmming skill, sorry to say that the strategy is not correct.
Try BB squeeze out method. You will get good % of returns.
Thanks for your understanding.
Hi
Can you combine all your strategies in one programm? Thanks
That is 5% profit on each trade. How many trades were there?
2 Trades
Thanks man. Amazing. Please wite one code with kucoin api. Im in iran and i cant use binance api.
Thanks mate. Best greetings to Iran 🇮🇷 I will see what I can do. I have covering another exchange on my to do list. Can't promise it will be Kucoin.
Loved the video and followed along while coding with you, got all the same results though my x axis doesn' t show dates. I then got up to the plt.scatter for marking Buy_Signal syntax I can't seem to get it to plot. @11:20 and 18:20
plt.scatter(df.index[df.Buy_Signal], df[df.Buy_Signal].Close, marker = '^', color='g')
in concatenate(*args, **kwargs).
The later return of buy and sell points works fine... just the scatter plot seems to be misbehaving.
Thanks for your kind words man. Well that could be a lot of reasons. Can you double check the df.Buy_Signal series? Did the np.where function create a boolean series? I think that might be the problem.
You should give it 10k dollars. And see how it would compound
No, that's a very very bad idea.
@@Algovibes not in real life! I mean 10k fake dollars on the python script. Just to see how it would do with compounding.
@@enzanto Ah, I actually did that in a lot of my other videos. Be invited to check out the Python for Finance and cryptobot playlist.!
@@Algovibes will do! Stumbled upon your channel last night. As I am in a learning period of both python, and stock market it is very interesting.
hi! thanks for the vid. I get this error everytime from the yfinance library. : yfinance\base.py", line 285, in history
df.index = df.index.tz_localize("UTC").tz_convert(
AttributeError: 'Index' object has no attribute 'tz_localize'
anyone ever seen this?
Welcome man. When exactly do you get this error?
@@Algovibes appreciate your answer!
when i put certain intervals in the yf.download function
@@thomazzzzi Just checked. I don't get this error. Can you just update yfinance?
Friend, I seem to get caught in the second line of code. I have all the packages, but it returns this error 'json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)'. Cheers
Yep, I had the same problem. Please check my newest community post. This should solve the error. Please don't forget to restart Jupyter after updating the yfinance lib.
Would be helpful if you can confirm this is solving your problem!
@@Algovibes Thanks for that, unfortunately no luck, same error. Even after restarting the computer. I googled it, and some are saying it's a problem with yahoo api, but strangely it works for others.
Please message me I would love to work on a project with you, if you are intrested