# authors: Affrin Sultana, Helin Wang, Shi Yan Wang and Pavel Levchenko
# January,2022
# import plotly.express as px
import plotly.graph_objects as go
import altair as alt
import pandas as pd
import numpy as np
import yfinance as yf
import pandas_datareader as pdr
import datetime
import warnings
alt.renderers.enable('altair_viewer')
[docs]def percent_change(stock_ticker, start_date, end_date):
"""
Calculates daily percentage change of a stock price within a given period of time
Parameters
----------
stock_ticker : string
Ticker of the stock such as 'AAPL', or 'AAPL MSFT SPY' for multiple tickers
start_date : string
Initial date for data extraction
end_date : string
Final date for stock analysis
Returns
--------
DataFrame
A data frame with dates and their corresponding stock price percentage changes.
Examples
--------
>>> percent_change('AAPL', '2017-01-01', '2017-01-10')
Price Change Percentage(%)
Date
2017-01-03 0.0000
2017-01-04 -0.1119
2017-01-05 0.3960
2017-01-06 1.5153
2017-01-09 2.4451
>>> percent_change('AAPL MSFT', '2017-01-01', '2017-01-10')
Price Change Percentage(%)
AAPL MSFT
Date
2017-01-03 0.0000 0.0000
2017-01-04 -0.1119 -0.4474
2017-01-05 0.3960 -0.4474
2017-01-06 1.5153 0.4155
2017-01-09 2.4451 0.0959
"""
# Assert ticker input value
ticker = yf.Ticker(stock_ticker)
if(ticker.info["regularMarketPrice"] == None):
raise NameError("You have entered an invalid stock ticker! Try again.")
# Assert start date input value
format = "%Y-%m-%d"
try: datetime.datetime.strptime(start_date, format)
except ValueError:
raise ValueError("You have entered an invalid start date! Try date formatted in YYYY-MM-DD.")
# Assert end date input value
try: datetime.datetime.strptime(end_date, format)
except ValueError:
raise ValueError("You have entered an invalid end date! Try date formatted in YYYY-MM-DD.")
# Assert end date is later than start date
format = "%Y-%m-%d"
if(datetime.datetime.strptime(end_date, format) < datetime.datetime.strptime(start_date, format)):
raise ValueError("You have entered an end date which is earlier than the start date! Try again.")
# Import original dataframe by giving stock ticker, start data and end date
data = yf.download(stock_ticker, start=start_date, end=end_date)
# Only Keep "Adj Close" Price for
data = data.drop(columns={'Open', 'High', 'Low', 'Adj Close', 'Volume'})
# Carry out calculation
for i in range(1,len(data)):
data.iloc[i,:] = (data.iloc[i,:] - data.iloc[0,:])/data.iloc[0,:]*100
data.iloc[0,:] = (data.iloc[0,:] - data.iloc[0,:])/data.iloc[0,:]*100
# Manipulate column name
data = data.rename(columns={"Close": "Price Change Percentage(%)"})
# Return result
return pd.DataFrame(data)
[docs]def profit_viz(stock_ticker, start_date , end_date, benchmark_ticker):
"""
Visualizes trend of a stock price change against the market benchmark within a given period of time
Parameters
----------
stock_ticker : string
Ticker of the stock such as 'AAPL'
start_date : string
Initial date for data extraction
end_date : string
Final date for stock analysis
benchmark_ticker : string
Ticker for benchmark comparison such as 'SP500'
Returns
--------
Line plots which shows percentage change in stock price and market performance over time
Examples
--------
>>> profit_viz('AAPL', '2015-01-01', '2021-31-12', 'SP500')
"""
ticker = yf.Ticker(stock_ticker)
bench_ticker = yf.Ticker(benchmark_ticker)
try:
# Assert ticker input value
if(ticker.info["regularMarketPrice"] == None):
raise NameError("You have entered an invalid stock ticker! Try again.")
# check data type of input
if type(stock_ticker) != str:
raise TypeError("stock_ticker should be of type string.")
# Assert benchmark ticker input value
if(bench_ticker.info["regularMarketPrice"] == None):
raise NameError("You have entered an invalid benchmark ticker! Try again.")
# check data type of input
if type(benchmark_ticker) != str:
raise TypeError("Bench Mark ticker should be of type string.")
# #check stock ticker and bench mark ticker are not same
# if stock_ticker is bench_ticker:
# raise NameError("Stock Mark ticker should not be same as Bench Ticker.")
# #check stock ticker is not empty
# if not stock_ticker or not bench_ticker:
# raise ValueError("'Tickers' cannot be empty")
# Assert start date input value
format = "%Y-%m-%d"
try: datetime.datetime.strptime(start_date, format)
except ValueError:
raise ValueError("You have entered an invalid start date! Try date formatted in YYYY-MM-DD.")
# Assert end date input value
try: datetime.datetime.strptime(end_date, format)
except ValueError:
raise ValueError("You have entered an invalid end date! Try date formatted in YYYY-MM-DD.")
# Assert end date is later than start date
format = "%Y-%m-%d"
if(datetime.datetime.strptime(end_date, format) < datetime.datetime.strptime(start_date, format)):
raise ValueError("You have entered an end date which is earlier than the start date! Try again.")
except (TypeError, ValueError, NameError) as err:
print(err)
raise
# Code to generate the visualization of profit
try:
stock_profit = percent_change(stock_ticker, start_date, end_date).reset_index()
benchmark_profit = percent_change(benchmark_ticker, start_date, end_date).reset_index()
profit_df = pd.merge(stock_profit, benchmark_profit, on="Date")
profit_df.rename({'Price Change Percentage(%)_x': 'Profit Percent Stock', 'Price Change Percentage(%)_y': 'Profit Percent Benchmark'}, axis=1, inplace=True)
# catch when dataframe is None
except AttributeError:
pass
#Checks if the datatype of data frame is correct
try:
isinstance(profit_df, pd.DataFrame)
except ValueError:
raise ValueError("profit_df is not a pandas dataframe.")
try:
isinstance(stock_profit, pd.DataFrame)
except ValueError:
raise ValueError("stock_profit couldnot be converted to a pandas dataframe.")
try:
isinstance(benchmark_profit, pd.DataFrame)
except ValueError:
raise ValueError("Benchmark_profit couldnot be converted to a pandas dataframe.")
# Code to plot the profit visualization
chart = alt.Chart(profit_df, title='Profit Percent trend of Stock vs Benchmark ticker').mark_line().transform_fold(
fold=['Profit Percent Stock', 'Profit Percent Benchmark'],
as_=['company', 'Profit Percent']
).encode(
x='Date:T',
y='Profit Percent:Q',
color=alt.Color('company:N', scale= alt.Scale(domain=['Profit Percent Stock','Profit Percent Benchmark'], range=['red', 'blue'])),
tooltip=[alt.Tooltip('Profit Percent Stock'),alt.Tooltip('Profit Percent Benchmark')]
)
return chart
[docs]def volume_change(stock_ticker, start_date, end_date):
"""
Calculates the daily trading volume change status of a stock within a given period of time
Parameters
----------
stock_ticker : string
Ticker of the stock such as 'AAPL'
start_date : string
Initial date for data extraction
end_date : string
Final date for stock analysis
Returns
--------
A data frame with dates and their corresponding trading volume and changes
Examples
--------
>>> volume_change('AAPL', '2021-01-01', '2022-01-01')
Date Volume Volume_Change
01-01-2021 1000 Nan
01-02-2021 2000 Increase
01-03-2021 3000 Increase
01-04-2021 2500 Decrease
...
12-31-2021 4000 Increase
01-01-2022 5000 Increase
"""
yf.pdr_override()
# Assert ticker value
ticker = yf.Ticker(stock_ticker)
if(ticker.info["regularMarketPrice"] == None):
raise NameError("You have entered an invalid stock ticker! Try again.")
# Assert date value
format = "%Y-%m-%d"
try: datetime.datetime.strptime(start_date, format)
except ValueError:
raise ValueError("You have entered an invalid start date! Try again.")
try: datetime.datetime.strptime(end_date, format)
except ValueError:
raise ValueError("You have entered an invalid end date! Try again.")
df = pdr.get_data_yahoo(stock_ticker, start=start_date, end=end_date)['Volume'].reset_index()
# Assert correct data fetched
try:
isinstance(df, pd.DataFrame)
except ValueError:
raise ValueError("Your input can't be converted to a pandas dataframe.")
df['Volume_dif'] = df['Volume'].diff().to_frame()
df["Volume_Change"] = np.select([df["Volume_dif"] > 0, df["Volume_dif"]<0], ["Increase", "Decrease"],
default = np.nan)
# Assert correct indicator values
for indicator in df["Volume_Change"]:
if(indicator != "Decrease" and indicator != "Increase" and indicator != "nan"):
raise ValueError("Incorrect Volume Change indicator")
return df[['Date', 'Volume', 'Volume_Change']]
[docs]def volume_viz(stock_ticker, start_date, end_date):
"""
Visualize the daily trading volume of a stock using bar plot within a given period of time
Parameters
----------
stock_ticker : string
Ticker of the stock such as 'AAPL'
start_date : string
Initial date for data extraction
end_date : string
Final date for stock analysis
Returns
--------
Interactive plot with overlay of line plots and bar plot
Examples
--------
>>> volume_viz('AAPL', '2021-01-01', '2022-01-01')
"""
try:
vdf = volume_change(stock_ticker, start_date, end_date)
# catch when dataframe is None
except AttributeError:
raise AttributeError("Invalid volume change input!")
vdf_increase = vdf.loc[vdf['Volume_Change']=='Increase']
vdf_decrease = vdf.loc[vdf['Volume_Change']=='Decrease']
fig = go.Figure()
fig.add_trace(go.Bar(x=vdf_increase['Date'], y=vdf_increase['Volume'],
base=0,
marker_color='green',
name='Volume Increase'))
fig.add_trace(go.Bar(x=vdf_decrease['Date'], y=vdf_decrease['Volume'],
base=0,
marker_color='red',
name='Volume Decrease'
))
return fig