Market Abuse Detection

In this task we need to retrieve stock data for Amazon and analyse some client data in order to find some suspicious behaviours.

Data description

• The file traders_data.csv includes some traders orders data. All those trades only reflect order submissions, which may have been filled or not.
• Amazon stock data should be retrieved from Yahoo for the months of February and March 2020 using the library pandas_datareader.

Task

We want to find traders which made suspicious orders. To be a suspicious orders we consider the following rules:

• The trader has submitted an order above the high price/below the low price for a given day of a stock
• The trader has submitted an order in a date when the stock was not traded

If any suspicious orders are found, we want to do the following analysis:

• If more than one trader is found, rank by number of suspicious orders per trader.
• Try to find if there is a correlation between the nationality of the trader and the tendency to make suspicious orders

What is evaluated

• Code logic/functionality (use of Pandas)
• Organisation of the code (python modules)
• Performance considerations
• Data processing/cleaning
• Documentation (mainly docstrings) Testing (unittests or pytest)