Backtrader, oh how we have missed you! It has been too long since the last article on this excellent platform. Hopefully today, we can make up for that neglect by using Backtrader as the engine for a stock screener. The platform is a perfect choice for a stock screener given how easy it is to create custom Indicators. Couple that with an already impressive library of built-in options and it becomes easy to screen for whether your favourite indicator is bullish or bearish.
Pre-Reading
Before we start, it is worth pointing out that this article builds on the shoulders of some other posts. More specifically, we will be using Alpha Vantage to provide data for the screener. As such, if you are interested in the mechanics of how we download the data, it would be worth looking at those articles first.
Even if not, at a bare minimum, you will need to sign up for an API key from Alpha Vantage. Again, if you are not sure how to do this, the same articles will cover those steps:
Scope
Moving onto scope, the code should allow users to download daily data for up to 500 instruments. The data will then be fed into Backtrader which will, in turn, run through the data and calculating indicator values. At the end of the run, we will create a report based on the values of the last bar of data to show if the indicators are bullish or bearish.
The idea is that you could then follow the example to add to or replace indicators that interest you.
Before Running
Make sure you find the line the following line in the code example and insert your API key.
|
1 |
Apikey = 'INSERT YOUR API KEY HERE' |
As you might imagine, the script will not work if you don’t!
Code
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 |
''' Author: www.backtest-rookies.com MIT License Copyright (c) 2019 backtest-rookies.com Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. ''' from alpha_vantage.timeseries import TimeSeries import pandas as pd import numpy as np import backtrader as bt from datetime import datetime import time # IMPORTANT! # ---------- # Register for an API at: # https://www.alphavantage.co/support/#api-key # Then insert it here. Apikey = 'INSERT YOUR API KEY HERE' def alpha_vantage_eod(symbol_list, compact=False, debug=False, *args, **kwargs): ''' Helper function to download Alpha Vantage Data. This will return a nested list with each entry containing: [0] pandas dataframe [1] the name of the feed. ''' data_list = list() size = 'compact' if compact else 'full' count = 0 total = len(symbol_list) for symbol in symbol_list: count += 1 print('\nDownloading: {}'.format(symbol)) print('Symbol: {} of {}'.format(count, total, symbol)) print('-'*80) # Submit our API and create a session alpha_ts = TimeSeries(key=Apikey, output_format='pandas') data, meta_data = alpha_ts.get_daily(symbol=symbol, outputsize=size) #Convert the index to datetime. data.index = pd.to_datetime(data.index) data.columns = ['Open', 'High', 'Low', 'Close','Volume'] if debug: print(data) data_list.append((data, symbol)) # Sleep to avoid hitting API limit print('Sleeping |', end='', flush=True) for x in range(12): print('=', end='', flush=True) time.sleep(1) print('| Done!') return data_list class TestStrategy(bt.Strategy): def __init__(self): self.inds = dict() self.inds['RSI'] = dict() self.inds['SMA'] = dict() for i, d in enumerate(self.datas): # For each indicator we want to track it's value and whether it is # bullish or bearish. We can do this by creating a new line that returns # true or false. # RSI self.inds['RSI'][d._name] = dict() self.inds['RSI'][d._name]['value'] = bt.indicators.RSI(d, period=14) self.inds['RSI'][d._name]['bullish'] = self.inds['RSI'][d._name]['value'] > 50 self.inds['RSI'][d._name]['bearish'] = self.inds['RSI'][d._name]['value'] < 50 # SMA self.inds['SMA'][d._name] = dict() self.inds['SMA'][d._name]['value'] = bt.indicators.SMA(d, period=20) self.inds['SMA'][d._name]['bullish'] = d.close > self.inds['SMA'][d._name]['value'] self.inds['SMA'][d._name]['bearish'] = d.close < self.inds['SMA'][d._name]['value'] def stop(self): ''' Called when backtrader is finished the backtest. Here we will just get the final values at the end of testing for each indicator. ''' # Assuming all symbols are going to have the same data on the same days. # If that is not the case and you are mixing assets from different classes, # regions or exchanges, then you might want to conisder adding an extra # column to the final results. print('{}: Results'.format(self.datas[0].datetime.date())) print('-'*80) results = dict() for key, value in self.inds.items(): results[key] = list() for nested_key, nested_value in value.items(): if nested_value['bullish'] == True or nested_value['bearish'] == True: results[key].append([nested_key, nested_value['bullish'][0], nested_value['bearish'][0], nested_value['value'][0]]) # Create and print the header headers = ['Indicator','Symbol','Bullish','Bearish','Value'] print('|{:^10s}|{:^10s}|{:^10s}|{:^10s}|{:^10s}|'.format(*headers)) print('|'+'-'*10+'|'+'-'*10+'|'+'-'*10+'|'+'-'*10+'|'+'-'*10+'|') # Sort and print the rows for key, value in results.items(): #print(value) value.sort(key= lambda x: x[0])# Sort by Ticker for result in value: print('|{:^10s}|{:^10s}|{:^10}|{:^10}|{:^10.2f}|'.format(key, *result)) # Create an instance of cerebro cerebro = bt.Cerebro() # Add our strategy cerebro.addstrategy(TestStrategy) # Download our data from Alpha Vantage. symbol_list = ['LGEN.L','LLOY.L','NG.L', 'BDEV.L'] data_list = alpha_vantage_eod( symbol_list, compact=True, debug=False) for i in range(len(data_list)): data = bt.feeds.PandasData( dataname=data_list[i][0], # This is the Pandas DataFrame name=data_list[i][1], # This is the symbol timeframe=bt.TimeFrame.Days, compression=1 ) #Add the data to Cerebro cerebro.adddata(data) print('\nStarting Analysis') print('-'*80) # Run the strategy cerebro.run() |
Commentary
We kick-off this script by downloading data for each ticker in our ticker list. For this version, we house them in a simple list. If your list of tickers is going to be large or maintained somewhere else, users might want to build on this by importing a ticker list from a file.
When we start downloading data, we loop through each item in the list and make a call to the API. Unfortunately, due to Alpha Vantage having strict API limits, we need to perform a long wait between each data download request. This results in the script waiting 12 seconds between download calls because we are limited to 5 API calls per minute.
For this reason, this screener is not really intended to be used on intra-day timeframes. It would be too slow unless your ticker list was quite small. Having said that, if you wish to swap out the data part with another service, then the rest of the code should still work. Another limitation is that Alpha Vantage’s free API is limited to 500 calls a day. Therefore, if you are thinking of intra-day screening, you would need to avoid making too many sweeps.
After we have all the data, it is just a matter of adding our favorite indicators to a dictionary. The dictionary is structured in such a way that we can conveniently make a report from it later. During __init__()we create that dictionary and add the indicators and create some “data feeds”. These new feeds signal whether the indicator is bullish or bearish. Creating a feed is as simple as creating a condition which returns TrueorFalse. That is one of the great things about Backtrader, it is really easy to create new data feeds.
The RSI example shows this on the line:
|
1 |
self.inds['RSI'][d._name]['bullish'] = self.inds['RSI'][d._name]['value'] > 50 |
As you can see, we are just checking if the RSI is over 50. If it is, then the data feed will be True, if not, it will be False.
For more information see: https://www.backtrader.com/docu/concepts/#almost-everything-is-a-data-feed
Adding Extra Indicators
So if you want to add extra indicators, you simply need to follow the same format of the examples
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
self.inds = dict() self.inds['RSI'] = dict() self.inds['SMA'] = dict() for i, d in enumerate(self.datas): # For each indicator we want to track it's value and whether it is # bullish or bearish. We can do this by creating a new line that returns # true or false. # RSI self.inds['RSI'][d._name] = dict() self.inds['RSI'][d._name]['value'] = bt.indicators.RSI(d, period=14) self.inds['RSI'][d._name]['bullish'] = self.inds['RSI'][d._name]['value'] > 50 self.inds['RSI'][d._name]['bearish'] = self.inds['RSI'][d._name]['value'] < 50 |
That means:
- Create a new entry in
self.indsand make it as adict()type. - As you loop through the data feeds, use the data feed name (
d._name) to create another new entry in the entry you just made. - Add the indicator at a
['value']key. - Create a check for whether the indicator is bullish or bearish. This should be a boolean value.
As long as you follow the correct format when adding new indicators during __init__()then the stop()method will not require any updates.
Running the Code
After running the Backtrader screener, you should have a table output which looks like this:
If you are checking a lot of instruments, you might want to consider running this as a nightly job after the markets have closed.
Hi- I am a bit new to python and was trying out the above code. For whatever reason I was printing out the oldest line of data 4/3/19 of the data set rather than the most recent. How would I adjust it to the most recent data set?