wassname/options_backtester
1
0
Fork 0
mirror of https://github.com/wassname/options_backtester.git synced 2026-07-26 13:28:06 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fixed imports. Run examples using data from Tiingo

Browse Source
This commit is contained in:
Juan Pablo Amoroso
2020-02-04 17:09:20 -03:00
parent 8c4f53194e
commit f92aa9cfac
8 changed files with 1702 additions and 84655 deletions
Download Patch File Download Diff File Expand all files Collapse all files
asset_backtester/__init__.py
+2
View File
@@ -1 +1,3 @@
from . import datahandler, charts
from .backtester import Backtest
from .portfolio import *
asset_backtester/backtester.py
+26 -30
View File
@@ -1,7 +1,7 @@
import pandas as pd
import pyprind
from portfolio import Portfolio
from .portfolio import Portfolio
class Backtest:
@@ -9,7 +9,6 @@ class Backtest:
self.schema = schema
self._portfolio = None
self._data = None
self.data_symbol = None
@property
def portfolio(self):
@@ -27,9 +26,8 @@ class Backtest:
@data.setter
def data(self, data):
self._data = data
self.data_symbol = df_symbol(data)
def run(self, initial_capital=1_000_000, periods='1', sma_months=None):
def run(self, initial_capital=1_000_000, periods=1, sma_months=None):
"""Runs a backtest and returns a dataframe with the daily balance"""
assert self._data is not None
assert self._portfolio is not None
@@ -41,9 +39,9 @@ class Backtest:
data_iterator = self._data.iter_dates()
first_day = self._data['date'].iloc[0]
last_day = self._data['date'].iloc[-1]
rebalancing_days = pd.date_range(first_day, last_day, freq=periods +
first_day = self._data['date'].min()
last_day = self._data['date'].max()
rebalancing_days = pd.date_range(first_day, last_day, freq=str(periods) +
'BMS').to_pydatetime() if periods is not None else []
bar = pyprind.ProgBar(data_iterator.ngroups, bar_char='█')
@@ -55,14 +53,10 @@ class Backtest:
index=[self._data.start_date - pd.Timedelta(1, unit='day')])
for date, data in data_iterator:
if date == self._data._data['date'][0]:
if date in rebalancing_days or date == first_day:
self._rebalance_portfolio(data)
self._update_balance(date, data)
if date in rebalancing_days:
self._rebalance_portfolio(data)
bar.update()
self.balance['% change'] = self.balance['capital'].pct_change()
@@ -73,12 +67,15 @@ class Backtest:
def _rebalance_portfolio(self, data):
"""Rebalances the portfolio so that the total money is allocated according to the given percentages"""
money_total = self.current_cash + self.current_capital
for asset in self._portfolio.assets:
asset_current = data[data['symbol'] == asset.symbol]
asset_price = asset_current[self.schema['Adj Close']].values[0]
query = '{} == "{}"'.format(self.schema['symbol'], asset.symbol)
asset_current = data.query(query)
asset_price = asset_current[self.schema['adjClose']].values[0]
qty = (money_total * asset.percentage) // asset_price
inventory_entry = self.inventory[self.inventory['symbol'] == asset.symbol]
inventory_entry = self.inventory.query(query)
self.inventory.drop(inventory_entry.index, inplace=True)
updated_asset = pd.Series([asset.symbol, asset_price, qty])
updated_asset.index = self.inventory.columns
@@ -93,10 +90,11 @@ class Backtest:
costs = []
for asset in self._portfolio.assets:
asset_current = data[data['symbol'] == asset.symbol]
inventory_asset = self.inventory[self.inventory['symbol'] == asset.symbol]
query = '{} == "{}"'.format(self.schema['symbol'], asset.symbol)
asset_current = data.query(query)
inventory_asset = self.inventory.query(query)
cost = asset_current[self.schema['Adj Close']].values[0]
cost = asset_current[self.schema['adjClose']].values[0]
qty = inventory_asset['qty'].values[0]
costs.append(cost * qty)
@@ -110,29 +108,27 @@ class Backtest:
'capital': money_total,
}, name=date)
self.balance = self.balance.append(row)
class df_symbol:
def __init__(self, data, sma_months=None):
def __init__(self, data, sma_months = None):
self.columns = data['symbol'].drop_duplicates(keep='first')
self.columns = data['symbol'].drop_duplicates(keep = 'first')
cols = pd.MultiIndex.from_product([self.columns.to_list(),data.columns[1:-1].to_list()])
df = pd.DataFrame(columns = cols, index = data['date'].unique())
cols = pd.MultiIndex.from_product([self.columns.to_list(), data.columns[1:-1].to_list()])
df = pd.DataFrame(columns=cols, index=data['date'].unique())
for col in cols:
symbol = data[data['symbol']==col[0]]
symbol = data[data['symbol'] == col[0]]
symbol = symbol.set_index('date')
df[col[0],col[1]] = symbol[col[1]]
df[col[0], col[1]] = symbol[col[1]]
self.data_symbol = df
def sma(self, sma_days):
df = pd.DataFrame(columns = self.columns)
df = pd.DataFrame(columns=self.columns)
for col in self.columns:
df[col] = self.data_symbol[col]['Adj Close'].rolling(sma_days, min_periods = 10).mean()
df[col] = self.data_symbol[col]['Adj Close'].rolling(sma_days, min_periods=10).mean()
return df
asset_backtester/stats/charts.py → asset_backtester/charts.py
+9 -10
View File
@@ -3,6 +3,7 @@
import altair as alt
import pandas as pd
def returns_chart(report):
# Time interval selector
time_interval = alt.selection(type='interval', encodings=['x'])
@@ -61,20 +62,18 @@ def monthly_returns_heatmap(report):
return chart
def historical_values(data_sma, data_symbol, asset_name):
asset_sma = pd.DataFrame(data_sma[asset_name])
asset_sma = asset_sma.rename(columns = {asset_name:'value'})
asset_sma['id'] = ['sma value']*(len(asset_sma.index))
asset_sma = asset_sma.rename(columns={asset_name: 'value'})
asset_sma['id'] = ['sma value'] * (len(asset_sma.index))
asset_sma = asset_sma.dropna()
asset_value = pd.DataFrame(data_symbol[asset_name]['Adj Close'])
asset_value= asset_value.rename(columns={'Adj Close' :'value'})
asset_value['id'] = ['Adj Close']*(len(asset_value.index))
asset_value = asset_value.rename(columns={'Adj Close': 'value'})
asset_value['id'] = ['Adj Close'] * (len(asset_value.index))
asset_value = asset_value.append(asset_sma)
asset_value['index'] = asset_value.index
plot = alt.Chart(asset_value).mark_line().encode(x='index:T',
y=alt.Y('value:Q'),
color='id'
)
return plot
plot = alt.Chart(asset_value).mark_line().encode(x='index:T', y=alt.Y('value:Q'), color='id')
return plot
asset_backtester/datahandler/schema.py
+4 -1
View File
@@ -3,7 +3,10 @@ class Schema:
Used to run validations and provide uniform access to fields in the data set.
"""
columns = ["symbol", "date", "open", "close", "high", "low", "volume", "Adj Close"]
columns = [
"symbol", "date", "open", "close", "high", "low", "volume", "adjClose", "adjHigh", "adjLow", "adjOpen",
"adjVolume", "divCash", "splitFactor"
]
def canonical():
"""Builder method that returns a `Schema` with default mappings"""
asset_backtester/examples/example.ipynb
+1660 -84550
View File
File diff suppressed because one or more lines are too long
asset_backtester/portfolio/asset.py
+1 -1
View File
@@ -5,4 +5,4 @@ class Asset:
self.percentage = percentage
def __repr__(self):
return "Asset(symbol={}, percentage={}, direction={})".format(self.symbol, self.percentage, self.direction)
return "Asset(symbol={}, percentage={})".format(self.symbol, self.percentage)
asset_backtester/statistics/__init__.py
-1
View File
@@ -1 +0,0 @@
from .charts import returns_chart, returns_histogram, monthly_returns_heatmap
asset_backtester/statistics/charts.py
-62
View File
@@ -1,62 +0,0 @@
"""Generates charts from a portfolio report"""
import altair as alt
def returns_chart(report):
# Time interval selector
time_interval = alt.selection(type='interval', encodings=['x'])
# Area plot
areas = alt.Chart().mark_area(opacity=0.7).encode(x='index:T',
y=alt.Y('accumulated return:Q', axis=alt.Axis(format='%')))
# Nearest point selector
nearest = alt.selection(type='single', nearest=True, on='mouseover', fields=['index'], empty='none')
points = areas.mark_point().encode(opacity=alt.condition(nearest, alt.value(1), alt.value(0)))
# Transparent date selector
selectors = alt.Chart().mark_point().encode(
x='index:T',
opacity=alt.value(0),
).add_selection(nearest)
text = areas.mark_text(
align='left', dx=5,
dy=-5).encode(text=alt.condition(nearest, 'accumulated return:Q', alt.value(' '), format='.2%'))
layered = alt.layer(selectors,
points,
text,
areas.encode(
alt.X('index:T', axis=alt.Axis(title='date'), scale=alt.Scale(domain=time_interval))),
width=700,
height=350,
title='Wealth over time')
lower = areas.properties(width=700, height=70).add_selection(time_interval)
return alt.vconcat(layered, lower, data=report.reset_index())
def returns_histogram(report):
bar = alt.Chart(report).mark_bar().encode(x=alt.X('% change:Q',
bin=alt.BinParams(maxbins=100),
axis=alt.Axis(format='%')),
y='count():Q')
return bar
def monthly_returns_heatmap(report):
resample = report.resample('M')['capital'].last()
monthly_returns = resample.pct_change().reset_index()
monthly_returns['capital'].iat[0] = resample.iloc[0] / report.iloc[0]['capital'] - 1
monthly_returns.columns = ['date', 'capital']
chart = alt.Chart(monthly_returns).mark_rect().encode(
alt.X('year(date):O', title='Year'), alt.Y('month(date):O', title='Month'),
alt.Color('mean(capital)', title='Return', scale=alt.Scale(scheme='redyellowgreen')),
alt.Tooltip('mean(capital)', format='.2f')).properties(title='Monthly Returns')
return chart