A Beginner’s Guide to NumPy, My 3 years Experience in Data Science

				
					# A Practical NumPy Project: Simple Stock Portfolio Analysis

# Step 1: Import the NumPy library.
# We always import it with the alias 'np' by convention.
import numpy as np

print("--- Step 1: Generating Sample Stock Data ---")

# Let's create a hypothetical dataset for daily stock prices.
# Imagine we are tracking 3 different stocks over 10 trading days.
# The `np.random.rand` function creates an array of random numbers
# between 0 and 1. We'll multiply and add to get more realistic prices.
# The shape is (10, 3) for 10 days and 3 stocks.
stock_prices = np.round(np.random.rand(10, 3) * 50 + 100, 2)

# `np.round()` is a handy function to round to a specific number of decimal places.
# We'll print the data so you can see what we're working with.
print("Daily Stock Prices (10 days for 3 stocks):\n", stock_prices)
print("-" * 50)

# Step 2: Calculating Daily Returns

# The daily return is the percentage change from one day to the next.
# We can calculate this using NumPy's slicing capabilities and vectorized operations.
# To do this, we'll take the prices from day 2 to the end and divide them
# by the prices from day 1 to the second to last day.
# We subtract 1 to get the percentage change.
# The result will be a (9, 3) array, as we lose the first day's data.

# `stock_prices[1:]` gives us all rows from the second row to the end.
# `stock_prices[:-1]` gives us all rows from the beginning up to the second to last day.
daily_returns = (stock_prices[1:] / stock_prices[:-1]) - 1

print("Daily Returns (as a percentage):\n", np.round(daily_returns * 100, 2))
print("-" * 50)

# Step 3: Calculating Key Metrics for the Portfolio

# Now, let's use some core NumPy functions to get meaningful insights from our data.
# We will use the 'axis' parameter to specify whether we want to perform
# the calculation along the rows (axis=0) or the columns (axis=1).

# Let's calculate the average price for each stock across all days.
# We use axis=0 because we want to average down the rows (each column represents a stock).
average_prices = np.mean(stock_prices, axis=0)
print("Average Price for each stock:", np.round(average_prices, 2))
print("-" * 50)

# Now, let's find the total cumulative return for each stock.
# We can sum up the daily returns.
total_returns = np.sum(daily_returns, axis=0)
print("Total Return for each stock (in percent):", np.round(total_returns * 100, 2))
print("-" * 50)

# Finally, let's calculate the standard deviation of the daily returns.
# Standard deviation is a measure of volatility. A higher number means more risk.
# We use axis=0 again to get the standard deviation for each stock.
volatility = np.std(daily_returns, axis=0)
print("Volatility (Standard Deviation) for each stock:", np.round(volatility, 4))
print("-" * 50)

# Step 4: Simple Portfolio Performance

# Let's imagine we invested a fixed amount in each stock.
initial_investment = np.array([1000, 1000, 1000]) # $1000 in each stock

# We can find the final value of our investment using vectorized multiplication.
# The first row of `stock_prices` is our initial price point.
final_value = initial_investment * (1 + total_returns)

print("Final value of initial $1000 investment in each stock:")
print(np.round(final_value, 2))