- Notebook Author: Trenton McKinney
- Course: DataCamp: Introduction to Python for Finance
- This notebook was created as a reproducible reference.
- The material is from the course
- I completed the exercises
- If you find the content beneficial, consider a DataCamp Subscription.
My Comments about this course¶
- Most of the course material is very basic so has not been included in the Notebook
- Great course for those with little or not Python experience
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
from numpy import NaN
from glob import glob
import re
pd.set_option('max_columns', 200)
pd.set_option('max_rows', 300)
pd.set_option('display.expand_frame_repr', True)
Data Files Location¶
- Most data files for the exercises can be found on the course site
- Other data files may be found in my DataCamp repository
Data File Objects¶
stocks1 = 'data/intro_to_python_for_finance/stock_data.csv'
stocks2 = 'data/intro_to_python_for_finance/stock_data2.csv'
sector_100 = 'data/intro_to_python_for_finance/sector.txt'
exercises = 'data/intro_to_python_for_finance/exercise_data.csv'
Intro to Python for Finance¶
Course Description
The financial industry is increasingly adopting Python for general-purpose programming and quantitative analysis, ranging from understanding trading dynamics to risk management systems. This course focuses specifically on introducing Python for financial analysis. Using practical examples, you will learn the fundamentals of Python data structures such as lists and arrays and learn powerful ways to store and manipulate financial data to identify trends.
Welcome to Python¶
This chapter is an introduction to basics in Python, including how to name variables and various data types in Python.
Welcome to Python for Finance!¶
Comments and Variables¶
Using variables to evaluate stock trends
$\text{Price to earning ratio} = \frac{\text{Maket Price}}{\text{Earnings per share}}$
price = 200
earnings = 5
pe_ratio = price/earnings
pe_ratio
Booleans in Python¶
Booleans are used to represent True or False statements in Python. Boolean comparisons include:
| Operators | Descriptions |
|---|---|
| > | greater than |
| >= | greater than or equal |
| < | less than |
| <= | less than or equal |
| == | equal (compare) |
| != | does not equal |
Lists¶
This chapter introduces lists in Python and how they can be used to work with data.
Lists¶
Nested Lists¶
List methods and functions¶
| Methods | Functions |
|---|---|
| All methods are functions | Not all functions are methods |
| List methods are a subset of built in functions in Python | |
| Used on an object | Requires an input of an object |
| prices.sort() | type(prices) |
- Functions take objects as inputs or are "passed" an object
- Methods act on objects
Arrays in Python¶
This chapter introduces packages in Python, specifically the NumPy package and how it can be efficiently used to manipulate arrays.
Arrays¶
Why use an array for financial analysis?
- Arrays can handle very large datasets efficiently
- Computationally memory efficient
- Faster calculations and analysis than lists
- Diverse functionality (many functions in Python packages)
- All dtypes in a numpy array are the same
- Each element of a python list keeps its dtype
Array operations¶
# Arrays - element-wise sum
array_A = np.array([1, 2, 3])
array_B = np.array([4, 5, 6])
array_A + array_B
# Lists - list concatenation
list_A = [1, 2, 3]
list_B = [4, 5, 6]
list_A + list_B
2D arrays and functions¶
Using arrays for analysis¶
Visualization in Python¶
In this chapter, you will be introduced to the Matplotlib package for creating line plots, scatter plots, and histograms.
Visualization in Python¶
Single Plot¶
df = pd.read_csv(stocks1)
df.head()
plt.plot(df.Day, df.Price, color='red', linestyle='--')
# Add x and y labels
plt.xlabel('Days')
plt.ylabel('Prices, $')
# Add plot title
plt.title('Company Stock Prices Over Time')
Multiple plots I¶
df = pd.read_csv(stocks2)
df.head()
# Plot two lines of varying colors
plt.plot(df.day, df.company1, color='red')
plt.plot(df.day, df.company2, color='green')
# Add labels
plt.xlabel('Days')
plt.ylabel('Prices, $')
plt.title('Stock Prices Over Time')
Multiple plots II¶
df[['company1', 'company2']].plot()
Scatterplots¶
plt.scatter(df.day, df.company1, color='green', s=0.1)
Histograms¶
- Tell the distribution of the data
- Uses in Finance
- Economic Indicators
- Stock Returns
- Commodity Prices
Why histograms for financial analysis?¶
- Is you data skewed?
- Is you data centered around the average?
- Do you have any abnormal data points (outliers) in your data?
Histograms and matplotlib.pyplot¶
import matplotlib.pyplot as plt
plt.hist(x=prices, bins=3)
plt.show()
Normalizing histogram data¶
import matplotlib.pyplot as plt
plt.hist(x=prices, bins=6, density=True)
plt.show()
- At times it's beneficial to know the relative frequency or the percentage of observations (rather than frequency counts)
Layering histograms on a plot¶
plt.hist(x=prices, bins=6, density=True)
plt.hist(x=prices2, bins=6, density=True)
plt.show()
Alpha: Changing transparency of histograms¶
plt.hist(x=prices, bins=6, density=True, alpha=0.5)
plt.hist(x=prices2, bins=6, density=True, alpha=0.5)
plt.show()
Adding a legend¶
plt.hist(x=prices, bins=6, density=True, alpha=0.5, label='Prices 1')
plt.hist(x=prices2, bins=6, density=True, alpha=0.5, label='Prices New')
plt.legend()
plt.show()
Exercises¶
Is data normally distributed?¶
plt.hist(df.company2, bins=100, ec='black')
plt.show()
Comparing two histograms¶
df_exercises = pd.read_csv(exercises)
df_exercises.head()
pd.DataFrame.hist¶
df_exercises.hist(bins=100, alpha=0.4, ec='black')
plt.show()
matplotlib.pyplt as plt¶
# Plot histogram of stocks_A
plt.hist(df_exercises.stock_A, bins=100, alpha=0.4, label='Stock A')
# Plot histogram of stocks_B
plt.hist(df_exercises.stock_B, bins=100, alpha=0.4, label='Stock B')
# Add the legend
plt.legend()
# Display plot
plt.show()
S&P 100 Case Study¶
In this chapter, you will get a chance to apply all the techniques you learned in the course on the S&P 100 data.
Introducing the dataset¶
Overall Review¶
- Python shell and scripts
- Variables and data types
- Lists
- Arrays
- Methods and functions
- Indexing and subsetting
- Matplotlib
S&P 100 Companies¶
Standard and Poor's S&P 100:
- made up of major companies that span multiple industry groups
- used to measure stock performance of large companies
S&P 100 Sectors¶
The Data¶
- EPS: earning per share
df = pd.read_csv(sector_100)
df.head()
df.tail()
Price to Earnings Ratio¶
$\text{Price to earning ratio} = \frac{\text{Maket Price}}{\text{Earnings per share}}$
- The dollar amount one can expect to invest in a company in order to receive one dollar of the company's earnings
- The ratio for valuing a company that measures its current share price relative to the per-share earnings
- In general, higher P/E ratio idicates higher growth expectations
Case Study Objective I:¶
Given
- List of data describing the S&P 100: names, prices, earnigns, sectors
Objective Part I
- Explore and analyze the S&P 100 data, specifically the P/E ratios of S&P 100 companies
Methods¶
- Step 1: examine the lists
- Step 2: Convert lists to arrays
- Step 3: Elementwise array operations
Project Explorations¶
Data¶
names = df.Name.values
prices = df.Price.values
earnings = df.EPS.values
sectors = df.Sector.values
type(names)
Lists¶
Stocks in the S&P 100 are selected to represent sector balance and market capitalization. To begin, let's take a look at what data we have associated with S&P companies.
Four lists, names, prices, earnings, and sectors, are available in your workspace.
Instructions
- Print the first four items in names.
- Print the name, price, earning, and sector associated with the last company in the lists.
# First four items of names
print(names[:4])
# Print information on last company
print(names[-1])
print(prices[-1])
print(earnings[-1])
print(sectors[-1])
Arrays and NumPy¶
NumPy is a scientific computing package in Python that helps you to work with arrays. Let's use array operations to calculate price to earning ratios of the S&P 100 stocks.
The S&P 100 data is available as the lists: prices (stock prices per share) and earnings (earnings per share).
Instructions
- Import the numpy as np.
- Convert the prices and earnings lists to arrays, prices_array and earnings_array, respectively.
- Calculate the price to earnings ratio as pe.
# Convert lists to arrays
prices_array = np.array(prices)
earnings_array = np.array(earnings)
# Calculate P/E ratio
pe = prices/earnings
pe[:10]
A closer look at sectors¶
Case Study Objective II:¶
Given
- Numpy arrays of data describing the S&P 100: names, prices, earnings, sectors
Objective Part II
- Explore and analyze sector-specific P/E ratios within companies of the S&P 100
Methods¶
- Step 1: Create a boolean filtering array
- Step 2: Apply filtering array to subset another array
- Step 3: Summarize P/E ratios
- Calculate the average and standard deviation of these sector-specific P/E ratios
Project Explorations¶
Filtering arrays¶
In this lesson, you will focus on two sectors:
- Information Technology
- Consumer Staples
numpy is imported as np and S&P 100 data is stored as arrays: names, sectors, and pe (price to earnings ratio).
Instructions 1/2
- Create a boolean array to determine which elements in sectors are 'Information Technology'.
- Use the boolean array to subset names and pe in the Information Technology sector.
# Create boolean array
boolean_array = (sectors == 'Information Technology')
# Subset sector-specific data
it_names = names[boolean_array]
it_pe = pe[boolean_array]
# Display sector names
print(it_names)
print(it_pe)
Instructions 2/2
- Create a boolean array to determine which elements in sectors are 'Consumer Staples'.
- Use the boolean array to subset names and pe in the Consumer Staples sector.
# Create boolean array
boolean_array = (sectors == 'Consumer Staples')
# Subset sector-specific data
cs_names = names[boolean_array]
cs_pe = pe[boolean_array]
# Display sector names
print(cs_names)
print(cs_pe)
Summarizing sector data¶
In this exercise, you will calculate the mean and standard deviation of P/E ratios for Information Technology and Consumer Staples sectors. numpy is imported as np and the it_pe and cs_pe arrays from the previous exercise are available in your workspace.
Instructions 1/2
Calculate the mean and standard deviation of the P/E ratios (it_pe) for the Industrial Technology sector.
# Calculate mean and standard deviation
it_pe_mean = np.mean(it_pe)
it_pe_std = np.std(it_pe)
print(it_pe_mean)
print(it_pe_std)
Instructions 2/2
- Calculate the mean and standard deviation of the P/E ratios (cs_pe) for the Consumer Staples sector.
# Calculate mean and standard deviation
cs_pe_mean = np.mean(cs_pe)
cs_pe_std = np.std(cs_pe)
print(cs_pe_mean)
print(cs_pe_std)
Plot P/E ratios¶
Let's take a closer look at the P/E ratios using a scatter plot for each company in these two sectors.
The arrays it_pe and cs_pe from the previous exercise are available in your workspace. Also, each company name has been assigned a numeric ID contained in the arrays it_id and cs_id.
Instructions
- Draw a scatter plot of it_pe ratios with red markers and 'IT' label.
- On the same plot, add the cs_pe ratios with green markers and 'CS' label.
- Add a legend to this plot.
- Display the plot.
it_id = np.arange(0, 15)
cs_id = np.arange(0, 12)
# Make a scatterplot
plt.scatter(it_id, it_pe, color='red', label='IT')
plt.scatter(cs_id, cs_pe, color='green', label='CS')
# Add legend
plt.legend()
# Add labels
plt.xlabel('Company ID')
plt.ylabel('P/E Ratio')
plt.show()
Notice that there is one company in the IT sector with an unusually high P/E ratio
Visualizating trends¶
Methods¶
- Step 1: Make a histogram of the P/E ratios
- Step 2:
- Identify the outlier P/E ratio
- Create a boolean array filter to subset this company
- Filter out this company information from the provided datasets
Project Explorations¶
Histogram of P/E ratios¶
To visualize and understand the distribution of the P/E ratios in the IT sector, you can use a histogram.
The array it_pe from the previous exercise is available in your workspace.
Instructions
- Selectively import the pyplot module of matplotlib as plt.
- Plot a histogram of it_pe with 8 bins.
- Add the x-label as 'P/E ratio' and y-label as 'Frequency'.
- Display the plot.
# Plot histogram
plt.hist(it_pe, bins=8, ec='black')
# Add x-label
plt.xlabel('P/E ratio')
# Add y-label
plt.ylabel('Frequency')
# Show plot
plt.show()
Identify the outlier¶
- Histograms can help you to identify outliers or abnormal data points. Which P/E ratio in this histogram is an example of an outlier?
A stock with P/E ratio > 50.
Name the outlier¶
You've identified that a company in the Industrial Technology sector has a P/E ratio of greater than 50. Let's identify this company.
numpy is imported as np, and arrays it_pe (P/E ratios of Industrial Technology companies) and it_names (names of Industrial Technology companies) are available in your workspace.
Instructions
- Identify the P/E ratio greater than 50 and assign it to outlier_price.
- Identify the company with P/E ratio greater than 50 and assign it to outlier_name.
# Identify P/E ratio within it_pe that is > 50
outlier_price = it_pe[it_pe > 50]
# Identify the company with PE ratio > 50
outlier_name = it_names[it_pe == outlier_price]
# Display results
print(f'In 2017 {outlier_name[0]} had an abnormally high P/E ratio of {round(outlier_price[0], 2)}.')
Certificate¶
