Skip to content

Instantly share code, notes, and snippets.

@shashankvemuri
Last active November 13, 2022 05:17
Show Gist options
  • Save shashankvemuri/4d04ab449122c29056a93aae4e45620a to your computer and use it in GitHub Desktop.
Save shashankvemuri/4d04ab449122c29056a93aae4e45620a to your computer and use it in GitHub Desktop.
all the code for the project
import yfinance as yf
import streamlit as st
import datetime
import talib
import ta
import pandas as pd
import requests
yf.pdr_override()
st.write("""
# Technical Analysis Web Application
Shown below are the **Moving Average Crossovers**, **Bollinger Bands**, **MACD's**, **Commodity Channel Indexes**, and **Relative Strength Indexes** of any stock!
""")
st.sidebar.header('User Input Parameters')
today = datetime.date.today()
def user_input_features():
ticker = st.sidebar.text_input("Ticker", 'AAPL')
start_date = st.sidebar.text_input("Start Date", '2019-01-01')
end_date = st.sidebar.text_input("End Date", f'{today}')
return ticker, start_date, end_date
symbol, start, end = user_input_features()
def get_symbol(symbol):
url = "http://d.yimg.com/autoc.finance.yahoo.com/autoc?query={}&region=1&lang=en".format(symbol)
result = requests.get(url).json()
for x in result['ResultSet']['Result']:
if x['symbol'] == symbol:
return x['name']
company_name = get_symbol(symbol.upper())
start = pd.to_datetime(start)
end = pd.to_datetime(end)
# Read data
data = yf.download(symbol,start,end)
# Adjusted Close Price
st.header(f"Adjusted Close Price\n {company_name}")
st.line_chart(data['Adj Close'])
# ## SMA and EMA
#Simple Moving Average
data['SMA'] = talib.SMA(data['Adj Close'], timeperiod = 20)
# Exponential Moving Average
data['EMA'] = talib.EMA(data['Adj Close'], timeperiod = 20)
# Plot
st.header(f"Simple Moving Average vs. Exponential Moving Average\n {company_name}")
st.line_chart(data[['Adj Close','SMA','EMA']])
# Bollinger Bands
data['upper_band'], data['middle_band'], data['lower_band'] = talib.BBANDS(data['Adj Close'], timeperiod =20)
# Plot
st.header(f"Bollinger Bands\n {company_name}")
st.line_chart(data[['Adj Close','upper_band','middle_band','lower_band']])
# ## MACD (Moving Average Convergence Divergence)
# MACD
data['macd'], data['macdsignal'], data['macdhist'] = talib.MACD(data['Adj Close'], fastperiod=12, slowperiod=26, signalperiod=9)
# Plot
st.header(f"Moving Average Convergence Divergence\n {company_name}")
st.line_chart(data[['macd','macdsignal']])
## CCI (Commodity Channel Index)
# CCI
cci = ta.trend.cci(data['High'], data['Low'], data['Close'], n=31, c=0.015)
# Plot
st.header(f"Commodity Channel Index\n {company_name}")
st.line_chart(cci)
# ## RSI (Relative Strength Index)
# RSI
data['RSI'] = talib.RSI(data['Adj Close'], timeperiod=14)
# Plot
st.header(f"Relative Strength Index\n {company_name}")
st.line_chart(data['RSI'])
# ## OBV (On Balance Volume)
# OBV
data['OBV'] = talib.OBV(data['Adj Close'], data['Volume'])/10**6
# Plot
st.header(f"On Balance Volume\n {company_name}")
st.line_chart(data['OBV'])
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment