polity 0.0.1

placeholder.

🐋 equities

Overview:

equities allows for easy access to the SEC's XBRL Financial Statement Dataset
Parsed data is stored locally and served to the user in pandas dataframes

The Dataset:

https://www.sec.gov/dera/data/financial-statement-data-sets.html

Install:

pip3 install equities

Donate:

Consider donating bitcoin to fund the future development of this project.

bitcoin wallet address: 3LU5MEaAXRJoCo6vx67g1Jj7qDFRKhMs5t

TUTORIAL:

The library consists of two central objects, Universe and Company.

Universe:

Building the Universe

We begin by initializing our universe and downloading our sec data packages.

from equities import Universe
u = Universe()

Essential Methods

To get the number of companies in the universe call: len(u)

To get a dataframe of XBRL metadata from of all companies in the universe call:

u.properties()

"CIK" numbers are the sec's official unique identifier for public companies. To get a full list of the cik numbers call:

u.ciks()

Accessing Companies

Universe objects are indexable by "CIK" integers. As an example, to access the first company in the universe call:

first_cik = universe.ciks()[0]
u[first_cik] # This returns an Company object.

Company:

A Company object should be thought of as an abstract representation of a real company. Every company must have an associated Universe of origin.

from equities import Company

Accessing the Financial Statements

Consider the first Company in our universe, universe[u.ciks()[0]]. It is a Company object.

c = u[u.ciks()[0]]

Dataframes of the company's financial statements over the universe in question is given by:

c.income()      # income statement dataframe

c.balance()     # Balancesheet dataframe

c.cash()        # Cash Flow Statement dataframe

c.equity()      # Consolidated Equity dataframe

Additional Company Details

To get the XBRL metadata for a given company as a pandas series call:

c.properties()

Example

I really want to demonstrate the beauty of this dataset since this is often difficult when looking at thousands of numeric datatables. Let's take a very naive peek by plotting various statements as a kind of stacked timeseries.

The following is a start to finish example of how one might plot the financial statements of the first three companies in the universe.

To perform this experiment, run the following:

from equities import test
test()

Here is the code that this function executes:

import pandas as pd
from equities import Universe, Company
import matplotlib.pyplot as plt

u = Universe()
u.build()

k,f,s = 'bar',(20,10),True
for cik in u.ciks()[:3]:

    u[cik].income().T.plot(
        kind=k,
        figsize=f,
        stacked=s)

    u[cik].cash().T.plot(
        kind=k,
        figsize=f,
        stacked=s)

    u[cik].balance().T.plot(
        kind=k,
        figsize=f,
        stacked=s)

plt.show()