Adjusted Returns including Inflation
Hi All! In the part 6 of Financial Analytics series, we learnt how to fetch historical financial data from multiple sources using Python. In this tutorial, we will learn the ways of considering inflation in a return series to obtain adjusted returns including inflation and implementing them using Python. We will be fetching data from quandl and yfinance in this tutorial, so I recommend you to read part 6 of Financial Analytics series first. Also, if you’re new to this series and want to learn Financial Analytics using Python, go to part 1 of Financial Analytics series to have a good head-start!
Adjusted Returns including Inflation – Formula
We have R’ = (1 + R)/(1 + I) – 1
Here, R = simple return series, I = inflation rate series, R’ = Return series with inflation
Step 1: Downloading Customer Price Index from Quandl
In [1]:
# Importing libraries
import pandas as pd
import quandl as qd
import yfinance as yf
In [2]:
key = 'key_comes_here'
In [3]:
#Key authentication - key to be fetched from https://www.quandl.com/account/profile after signing up
qd.ApiConfig.api_key = key
In [4]:
#Fetching CPI - Customer Price Index for USA from quandl
USA_CPI = qd.get(dataset='RATEINF/CPI_USA',
start_date='2008-04-01',
end_date='2018-03-31')
In [5]:
#Checking head of the dataframe
USA_CPI.head()
Out[5]:
In [6]:
# Renaming Value column to CPI
# Calling rename function on USA_CPI and passing dictionary to columns having keys as old column names and values as new column names
USA_CPI.rename(columns = {'Value': 'CPI'}, inplace = True)
USA_CPI.head()
Out[6]:
Step 2: Fetching stocks data from yfinance
In [7]:
# Downloading MSFT data from yfinance from 1st April 2008 to 31st March 2018
msftStockData = yf.download( 'MSFT',
start = '2008-04-01',
end = '2018-03-31',
progress = False)
In [8]:
# Checking what's in there the dataframe by loading first 5 rows
msftStockData.head()
Out[8]:
Step 3: Making a date dataframe
In [9]:
dates = pd.DataFrame(index=pd.date_range(start='2008-04-01',
end='2018-03-31'))
dates.head()
Out[9]:
Step 4: Joining dates and stock data
In [10]:
# Joining by calling join function on dates dataframe.
# In how parameter, we specify which join we want to perform
# how = 'left' performs left join in which all rows of dates are fetched and matching rows from msftStockData is returned
# Unmatched values have NaN or Not a Number
# We are only interested in Adjusted close column, so we will include only that column from the right table
Date_Stock = dates.join(msftStockData[['Adj Close']], how = 'left')
Date_Stock.head(10)
Out[10]:
In [11]:
# In order to handling missing values, let's use missing value imputation
# Importing simpleImputer class from sklearn.impute library
from sklearn.impute import SimpleImputer
#Inititalizing the object of SimpleImputer class
data_Imputer = SimpleImputer()
#Fitting and transforming the data using the data_Imputer object and then wrapping it around pd.DataFrame function to convert it into a dataframe from numpy array.
imputedData = pd.DataFrame(data_Imputer.fit_transform(Date_Stock))
imputedData.head(10)
Out[11]:
In [12]:
# Replacing values from imputedData to Adj Close column of Date_Stock
Date_Stock['Adj Close'] = imputedData[0].values
Date_Stock.head(10)
Out[12]:
In [13]:
# Selecting end of month rows only to do Month on Month analysis
Date_Stock = Date_Stock.asfreq('M')
Date_Stock.head(10)
Out[13]:
Step 5: Joining Stock and CPI data
In [14]:
# Left joining stock and USA inflation (Customer Price Index) data
Stock_Inflation = Date_Stock.join(USA_CPI, how = 'left')
Stock_Inflation.head()
Out[14]:
Step 6: Calculating simple returns and inflation rate
In [15]:
#Calculating simple returns by pct_change() function of the dataframe
Stock_Inflation['Simple Return'] = Stock_Inflation['Adj Close'].pct_change()
Stock_Inflation.head()
Out[15]:
In [16]:
#Calculating inflation rate by pct_change() function of the dataframe
Stock_Inflation['Inflation rate'] = Stock_Inflation['CPI'].pct_change()
Stock_Inflation.head()
Out[16]:
Step 7: Considering inflation rate in returns
In [17]:
# Applying the formula R' = (1 + Simple Return)/(1 + Inflation rate) - 1
Stock_Inflation['Adj Return'] = ((1 + Stock_Inflation['Simple Return'])/(1 + Stock_Inflation['Inflation rate'])) - 1
Stock_Inflation.head()
Out[17]:
Step 8: Seeing the results on line charts
In [18]:
# importing matplotlib library
import matplotlib.pyplot as plt
In [19]:
# Seeing simple return on line chart
Stock_Inflation['Simple Return'].plot(figsize = (14,10))
Out[19]:
In [20]:
# Seeing adjusted return on line chart
Stock_Inflation['Adj Return'].plot(figsize = (14,10))
Out[20]:
In [21]:
# Seeing inflation rate on line chart
Stock_Inflation['Inflation rate'].plot(figsize = (14,10))
Out[21]:
So guys, in this tutorial, we learnt how to obtain adjusted return – considering inflation rate in simple returns.
Stay tuned for next tutorial on this Financial Analytics series. Don’t forget to subscribe to our YouTube channel.