quickData

Installation

Prerequisites

Python 3.6 +
lxml
requests
fuzzy_wuzzy

Note earlier versions of python can be used if all f-strings in dataCreator.py are refactored to normal string concatenations. i.e. change some_function(f'example {string}') to some_function('example ' + string)

Installation Steps

To install simply run pip install -r requirements.txt, or alternatively:

Download source files either manually or via git clone https://github.com/nikodevv/quickData
Install requirements using pip

	pip install lxml
	pip install requests
	pip install fuzzywuzzy

API

Simply create an instance of Filings with a given company's CIK, and a Python object containing all of the company's financial statements will be generated. For example Snapchat's CIK is 1564408 (Snap Inc.), so a Filings object would be created as follows:

snapFilings = Filings(1564408)

Example (generating time series data of a financial statement)

To access the data in any SEC-filing table (example: Snap Inc's 2017 Income Statement), just call

custom_table = [] 
# custom_table[0] will be the row labels of income statement
custom_table.append(snapFiligns.row_labels['income'])
# custom_table[1] will be the values
custom_table.append(snapFilings.full_dict['2017Q3'][income])

print(custom_table[0][1] + ':' + custom_table[1][1])
# Revenue: 207937

print(custom_table[0][9] + ':' + custom_table[1][9])
# Loss from operations: -461827

Note: row_labels will not be an exact copy of the orignial statement's table rows because quickData creates and removes certain rows for the purposes of consolidating multiple time periods which may utilize inconsistent reporting methodologies.

Explanation of example

The company's time-series filings are stored in two objects. The first is a dictionary of lists containing the row labels (snapFilings.row_labels) of all 3 financial statements. These labels are similiar to those found in the original SEC filings. Each financial statement can be accessed by one of three keys: income returns income statement line items, balance returns balance sheet line items, and cfs returns cashflow statement line items (line items refers to accounts). For example, the row labels of Snapchat's quickData income statement can be accessed as follows:

snapFilings.row_labels['income']
> ['Income Statement [Abstract]', 'Revenue', 'Costs and expenses', 'Cost of revenue', 'Research and development', 'Sales and marketing', 'General and administrative', ...]

The row_labels are in 1:1 correspondace with the data columns corresponding to each time period. These data columns are called by snapFilings.full_dict. Together, row_labels and full_dict can create time-series financial statements.

Here is an example of what Snap Inc.'s full_dict looks like:

# snapFilings.full_dict = 
	{'2017Q1': {
		'balance': [...]
		'income' : [...]
		'cfs': [...]
		}
	'2017Q2': {
		'balance': [...]
		'income' : [...]
		'cfs': [...]
		}
	'2017Q1' : ...
	}

Where 2017Q1 refers to the first financial quarter of Snap Inc's 2017 financial year. Annual filings keys are of the following form: '2017FY'.

Disclaimer

The author assumes no responsibility or liability for any errors, inaccuracies, or omissions in the data generated or output by quickData, nor any responsibility or liability for investment or business descicions made on said data. The information provided by quickData is provided on an “as is” basis with no guarantees of completeness, accuracy, usefulness or timeliness.

nikodevv / quickData