pandas-ta/examples/AIExample.ipynb

Strategy Analysis with Pandas TA and AI/ML

• This is a Work in Progress and subject to change!
• Contributions are welcome and accepted!
• Examples below are for educational purposes only.
• NOTE: The watchlist module is independent of Pandas TA. To easily use it, copy it from your local pandas_ta installation directory into your project directory.

Required Packages

Uncomment the packages you need to install or are missing

In [1]:

#!pip install numpy
#!pip install pandas
#!pip install mplfinance
#!pip install pandas-datareader
#!pip install requests_cache
#!pip install alphaVantage-api # Required for Watchlist

In [2]:

%pylab inline
import datetime as dt
import random as rnd
from sys import float_info as sflt

import numpy as np
import pandas as pd
pd.set_option('max_rows', 100)
pd.set_option('max_columns', 20)

import mplfinance as mpf
import pandas_ta as ta

from watchlist import colors, Watchlist # Is this failing? If so, copy it locally. See above.

print(f"Numpy v{np.__version__}")
print(f"Pandas v{pd.__version__}")
print(f"mplfinance v{mpf.__version__}")
print(f"Pandas TA v{ta.version}")
%matplotlib inline

Populating the interactive namespace from numpy and matplotlib
Numpy v1.19.2
Pandas v1.1.0
mplfinance v0.12.7a0
Pandas TA v0.2.28b0

MISC Functions

In [3]:

def recent_bars(df, tf: str = "1y"):
    # All Data: 0, Last Four Years: 0.25, Last Two Years: 0.5, This Year: 1, Last Half Year: 2, Last Quarter: 4
    yearly_divisor = {"all": 0, "10y": 0.1, "5y": 0.2, "4y": 0.25, "3y": 1./3, "2y": 0.5, "1y": 1, "6mo": 2, "3mo": 4}
    yd = yearly_divisor[tf] if tf in yearly_divisor.keys() else 0
    return int(ta.RATE["TRADING_DAYS_PER_YEAR"] / yd) if yd > 0 else df.shape[0]

Collect some Data

In [4]:

tf = "D"
tickers = ["SPY", "QQQ", "AAPL", "TSLA"]
watch = Watchlist(tickers, tf=tf, timed=True)
watch.strategy = ta.CommonStrategy
watch.load(tickers, analyze=True, verbose=False)

[!] Loading All: SPY, QQQ, AAPL, TSLA
[i] Loaded['D']: SPY_D.csv
[i] Runtime: 845.8352 ms (0.8458 s)
[i] Loaded['D']: QQQ_D.csv
[i] Runtime: 882.0416 ms (0.8820 s)
[i] Loaded['D']: AAPL_D.csv
[i] Runtime: 1685.3232 ms (1.6853 s)
[i] Loaded['D']: TSLA_D.csv
[i] Runtime: 1306.2404 ms (1.3062 s)

Select an Asset

In [5]:

ticker = tickers[0]
print(f"{ticker} {watch.data[ticker].shape}\nColumns: {', '.join(list(watch.data[ticker].columns))}")

SPY (5265, 10)
Columns: open, high, low, close, volume, SMA_10, SMA_20, SMA_50, SMA_200, VOL_SMA_20

Trim it

In [6]:

duration = "1y"
recent = recent_bars(watch.data[ticker], duration)
asset = watch.data[ticker].copy().tail(recent)

Create a Trend

In [7]:

# Example Long Trends
# long = ta.sma(asset.close, 10) < ta.sma(asset.close, 20) # SMA(10) > SMA(20)
long = ta.ema(asset.close, 8) > ta.ema(asset.close, 21) # EMA(8) > EMA(21)
# long = ta.increasing(ta.ema(asset.close, 50))
# long = ta.macd(asset.close).iloc[:,1] > 0 # MACD Histogram is positive

asset.ta.ema(length=8, append=True)
asset.ta.ema(length=21, append=True)
asset.ta.ema(length=50, append=True)
asset[asset.columns[5:]].tail()

Out [7]:

	SMA_10	SMA_20	SMA_50	SMA_200	VOL_SMA_20	EMA_8	EMA_21	EMA_50
date
2020-09-28	331.181	336.9395	334.8642	310.30500	86281647.00	330.408828	334.010756	331.721246
2020-09-29	330.401	336.0925	335.0252	310.38120	85553267.55	330.844644	333.861596	331.746687
2020-09-30	330.008	335.2070	335.2228	310.46905	88007358.10	331.743612	333.955087	331.869955
2020-10-01	330.128	334.1740	335.4264	310.55675	88965293.60	332.920587	334.235534	332.072702
2020-10-02	330.447	333.5965	335.6440	310.62810	86036292.75	333.124901	334.199576	332.142007

Calculate Trend Returns from the long trend

In [8]:

trendy = asset.ta.trend_return(trend=long, cumulative=True, trade_offset=-1, append=True)
trendy.tail() # Third Column is the long trend; binary sequences

Out [8]:

	CLTR	TR_LOGRET	CLTR_Trends	CLTR_Trades
date
2020-09-28	0.0	0.0	0	0
2020-09-29	0.0	0.0	0	0
2020-09-30	0.0	0.0	0	0
2020-10-01	0.0	0.0	0	0
2020-10-02	0.0	0.0	0	0

Simple Price Chart

In [9]:

extime = ta.get_time(to_string=True)
chart_ = asset[["close", "EMA_8", "EMA_21", "EMA_50"]]
chart_.plot(figsize=(16, 10), color=colors("BkGrOrRd"), title=f"{ticker} {extime}", grid=True)

Out [9]:

<AxesSubplot:title={'center':'SPY Wednesday December 23, 2020, NYSE: 8:37:29, Local: 12:37:29 PST, Day 358/365 (98.0%)'}, xlabel='date'>

Trend Returns and Cumulative Trend Returns

In [10]:

cltr = trendy.iloc[:,0]
tr = trendy.iloc[:,1]

trendy.iloc[:,:2].plot(figsize=(16, 3), color=colors("BkBl"))
cltr.plot(figsize=(16, 3), kind="area", stacked=False, color=colors("SvGy")[0], alpha=0.25, grid=True)

Out [10]:

<AxesSubplot:xlabel='date'>

Total Return

In [11]:

capital = 10000

total_return = cltr.cumsum() * capital
positive_return = total_return[total_return > 0]
negative_return = total_return[total_return <= 0]
trdf = pd.DataFrame({"tr+": positive_return, "tr-": negative_return})
trdf.plot(figsize=(16, 5), color=colors(), kind="area", stacked=False, alpha=0.25, grid=True)

Out [11]:

<AxesSubplot:xlabel='date'>

Long and Short Trends

In [12]:

long_trend = (trendy.iloc[:,-2] > 0).astype(int)
short_trend = (1 - long_trend).astype(int)

long_trend.plot(figsize=(16, 0.85), kind="area", stacked=True, color=colors()[0], alpha=0.25)
short_trend.plot(figsize=(16, 0.85), kind="area", stacked=True, color=colors()[1], alpha=0.25)

Out [12]:

<AxesSubplot:xlabel='date'>

Entries & Exits

In [13]:

entries = (trendy.iloc[:,-1] > 0).astype(int) * asset.close
entries[entries < 0.0001] = np.nan
entries.name = "Entry"


exits = (trendy.iloc[:,-1] < 0).astype(int) * asset.close
exits[exits < 0.0001] = np.nan
exits.name = "Exit"

total_trades = trendy.iloc[:,-1].abs().sum()
print(f"Total Trades: {total_trades}")

all_trades = trendy.iloc[:,-1].copy().fillna(0)
all_trades = all_trades[all_trades != 0]

trades = pd.DataFrame({"Signal": all_trades, entries.name: entries.dropna(), exits.name: exits.dropna()})
trades

Out [13]:

Total Trades: 6

	Signal	Entry	Exit
date
2019-10-31	1	303.33	NaN
2020-02-24	-1	NaN	322.42
2020-04-08	1	274.03	NaN
2020-06-26	-1	NaN	300.05
2020-06-29	1	304.46	NaN
2020-09-10	-1	NaN	333.89

Chart Display Strings

In [14]:

first_date, last_date = asset.index[0], asset.index[-1]
f_date = f"{first_date.day_name()} {first_date.month}-{first_date.day}-{first_date.year}"
l_date = f"{last_date.day_name()} {last_date.month}-{last_date.day}-{last_date.year}"
last_ohlcv = f"Last OHLCV: ({asset.iloc[-1].open}, {asset.iloc[-1].high}, {asset.iloc[-1].low}, {asset.iloc[-1].close}, {int(asset.iloc[-1].volume)})"
ptitle = f"\n{ticker} [{tf} for {duration}({recent} bars)] from {f_date} to {l_date}\n{last_ohlcv}\n{extime}"

Trade Chart

In [15]:

# chart = asset["close"] #asset[["close", "SMA_10", "SMA_20", "SMA_50", "SMA_200"]]
# chart = asset[["close", "SMA_10", "SMA_20"]]
chart = asset[["close", "EMA_8", "EMA_21", "EMA_50"]]
chart.plot(figsize=(16, 10), color=colors("BkGrOrRd"), title=ptitle, grid=True)
entries.plot(figsize=(16, 10), color=colors("FcLi")[1], marker="^", markersize=12, alpha=0.8)
exits.plot(figsize=(16, 10), color=colors("FcLi")[0], marker="v", markersize=12, alpha=0.8, grid=True)

Out [15]:

<AxesSubplot:title={'center':'\nSPY [D for 1y(252 bars)] from Friday 10-4-2019 to Friday 10-2-2020\nLast OHLCV: (331.7, 337.0126, 331.19, 333.84, 89431112)\nWednesday December 23, 2020, NYSE: 8:37:29, Local: 12:37:29 PST, Day 358/365 (98.0%)'}, xlabel='date'>

AI Analysis

In [ ]: