How to Calculate Simple Buy And Hold Strategy in Python

This Python script loads historical stock or index data (like the Nifty index) from a CSV file, processes it to calculate daily logarithmic returns, and visualizes both the raw prices and derived returns using plots and histograms. It then computes cumulative returns to simulate the performance of a long-term investment. Through the use of pandas, numpy, and matplotlib, it offers both numerical insights and clear visualizations to help analyze the asset’s behavior over time.

1.Importing Libraries

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

pandas (pd) is used for data manipulation and analysis.
numpy (np) adds support for numerical computations.
matplotlib.pyplot (plt) is used for plotting graphs.

2. Reading the Data

df = pd.read_csv("sample.csv", parse_dates=["Date"], index_col="Date")
df
df.info()

Loads sample.csv into a DataFrame df.
parse_dates=["Date"]: converts the “Date” column to datetime format.
index_col="Date": sets the “Date” column as the index for time series analysis.
df by itself would display the DataFrame.
df.info() gives a quick summary: datatypes, non-null counts, and memory usage.

3. Plotting the Data

df.plot(figsize=(12, 8), title="Sample Nifty Instrument", fontsize=12)
plt.show()

Plots all numerical columns in df.
The plot size is 12×8 inches, with a title and readable font.
plt.show() renders the plot.

4. Calculating Logarithmic Returns

df["returns"] = np.log(df.div(df.shift(1)))

Computes logarithmic returns for each column.
df.shift(1): shifts values down by one row (previous day’s price).
df.div(...): element-wise division (today’s price / yesterday’s).
np.log(...): takes the natural logarithm to compute returns.

5. Cleaning Data

df.dropna(inplace=True)

Removes any rows with NaN values (likely from the first row after .shift(1)).

6. Visualizing Returns

df["returns"].hist(bins=100, figsize=(12,8))
plt.title("Nifty Instrument")
plt.show()

Plots a histogram of daily returns with 100 bins to show distribution shape.

7. Cumulative Return Calculations

np.exp(df.returns.sum())

Calculates the cumulative return over time using log returns.

df.price[-1] / df.price[0]

Alternative method: compares final price to initial price (only works if price column exists!).

df.returns.cumsum().apply(np.exp)

Cumulative sum of log returns and converted back to normal scale using exp().

8. Adding and Plotting Cumulative Returns

df["creturns"] = df.returns.cumsum().apply(np.exp)
df.creturns.plot(figsize=(12,8), title="Nifty Instruments Hold", fontsize=12)
plt.show()

Adds a new column creturns for cumulative returns.
Plots the performance over time—like tracking your investment value if you held the asset throughout.

Example: full program to calculate stock returns

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt 
df = pd.read_csv("sample.csv", parse_dates=["Date"], index_col="Date")
df
df.info()
df.plot(figsize=(12, 8), title="Sample Nifty Instrument", fontsize=12)
plt.show()
df["returns"]=np.log(df.div(df.shift(1)))
df
df.dropna(inplace=true)
df
df["returns"].hist(bins=100,figsize=(12,8))
plt.title("Nifty Instrument ")
plt.show()
np.exp(df.returns.sum())
df.price[-1]/df.price[0]
df.returns.cumsum().apply(np.exp)

df["creturns"]=df.returns.cusum().apply(np.exp)
df
df.creturns.plot(figsize=(12,8) , title="nifty instruments hold" , fontsize=12)
plt.show()