Goal: Automatically generate an analysis for the company to review before moving forward with its quarterly report.
Data is loaded from csv in the following format:
This example is a small fraction of the data and only contains information from one location in a small range of dates (Around 11,000 transactions). In the full implementation, data will also be analyzed according to location to provide geo-specific insight into the company's success in relation to location.
The current repository only contains the foundation of the code that will be used to generate a full report. It is centered around detecting potential delinquent credit card charges or refunds.
This analysis flags and provides information on potentially misreported financial data, including outstanding charges, erroneous refunds, and unexplained jumps in the timing of charges and refunds. This generates a DataFrame such as the one below:
Additionally, the public code provides a simple report on the users with outstanding charges:
In future versions of the report, data visualizations will be generated to compare location as well as to summarize the data extracted
from this foundational EDA.