Assessment Index Modelling

The instruction of running the project: How to run the project

This Code Assessment simulates the scenario of building an index model based on a given guideline. The rules provided below describe a simple stock index and for this assessment the objective is to build a model that can calculate such index values.

The idea is not only that you can show us what you've got, but also that you can get a first impression of what it means to build an index model.

The required input prices can be found at data_sources\stock_prices.csv.

You can also find the correct (rounded) index level at data_sources\index_level_results_rounded.csv. Please note these values cannot be used as an input but solely act as a reference for you to verify your model. For the evaluation of your model we will use the high precision index values your model calculates.

Index Notes

• The index is a stock index made up of imaginary stocks.
• There are no further corporate actions to consider here
• The index doesn't resemble any real existing index.
• The model should be able to calculate the index levels based on the rules below.
• All provided prices are total return.
• All companies got the same amount of shares outstanding.

Index Rules

• The index is a total return index.
• The index universe consists of all stocks from "Stock_A" to including "Stock_J".
• Every first business day of a month the index selects from the universe the top three stocks based on their market capitalization, based on the close of business values as of the last business day of the immediately preceding month.
• The selected stock with the highest market capitalization gets assigned a 50% weight, while the second and third each get assigned 25%.
• The selection becomes effective close of business on the first business date of each month.
• The index starts with a level of 100.
• The index start date is January 1st 2020.
• The index business days are Monday to Friday.
• There are no additional holidays.

Setup

We've already provided a __main__.py for this project and there should be no need for you to modify it. As you will see the main file expects an index model which can be found at index_model/index.py. This is the model we expect you to build. It should then ultimately be able to export the calculated values to a csv file.

If you want to use additional packages feel free to do so. The actual business logic must be part of your repo, but we actively encourage you to use packages that you're already familiar with. Please make sure to update the requirements.txt file accordingly.

How to Run the Project

Run the following commands from the repository root.

1. Create a virtual environment

Windows PowerShell:

python -m venv .venv
.\.venv\Scripts\python.exe -m pip install --upgrade pip

macOS/Linux:

python3 -m venv .venv
./.venv/bin/python -m pip install --upgrade pip

2. Install dependencies

Windows PowerShell:

.\.venv\Scripts\python.exe -m pip install -r requirements.txt

macOS/Linux:

./.venv/bin/python -m pip install -r requirements.txt

3. Run the model

Windows PowerShell:

.\.venv\Scripts\python.exe __main__.py

macOS/Linux:

./.venv/bin/python __main__.py

The model reads prices from data_sources/stock_prices.csv and exports the calculated index levels to:

exported_data/export.csv

The rounded reference values in data_sources/index_level_results_rounded.csv can be used to verify the calculated output, but must not be used as model input.

Note

1. Parameter of the model are stored as variables at the beginning of index_model/index.py. Modify it when introducing new index rules.
2. The input parameter start_date of IndexModel.calc_index_level in the code is treated as the Start Date of the index as I assume the model is used for backtest based on the variable names in the pre-written __main__.py. In current __main__.py, the start date is January 1st 2020. Therefore, if a custom start_date is being used, it needs to be the first business day of a month.