Python for backtesting
Budget: ₹100 – ₹400 INR
I have several raw financial data sets—daily price feeds, transaction logs, and balance-sheet snapshots—sitting in CSV format. The task is to load these files into Python (think pandas and NumPy), clean and merge them where necessary, then dig into the numbers to pinpoint meaningful trends.
I need clear, reproducible insight rather than black-box output, so please organise the work in a Jupyter Notebook that:
• shows each transformation step,
• highlights any assumptions or filters applied, and
• concludes with well-labelled visualisations (matplotlib or seaborn are fine) that make the identified trends obvious at a glance.
A short written summary of the findings—key upward or downward movements, cyclical patterns, and anything unusual worth flagging—will round out the delivery. I’ll provide sample files and a data dictionary once we start; you simply hand back the notebook, cleaned data export, and the summary report.
Everything must run on Python 3.x with no proprietary libraries beyond the standard scientific stack so it can drop straight into my existing workflow.
I need clear, reproducible insight rather than black-box output, so please organise the work in a Jupyter Notebook that:
• shows each transformation step,
• highlights any assumptions or filters applied, and
• concludes with well-labelled visualisations (matplotlib or seaborn are fine) that make the identified trends obvious at a glance.
A short written summary of the findings—key upward or downward movements, cyclical patterns, and anything unusual worth flagging—will round out the delivery. I’ll provide sample files and a data dictionary once we start; you simply hand back the notebook, cleaned data export, and the summary report.
Everything must run on Python 3.x with no proprietary libraries beyond the standard scientific stack so it can drop straight into my existing workflow.