Forecast the Account Balance:
  • Input for Learning/Training: Transaction set in a csv file. Relevant parameters includes the Date, Transaction details which is a text, Payer/Payee, Amount of Debit/Credit and Balance
  • Output from Learning/Training:  A model that can forecast the following:
    • Cash Balance in future – at least with some reliability for the next week/month and up to 6 future months
    • Stretch goal – Can forecast the cash flow with some reliability for the next week/month and up to 6 future months. i.e. it can tell how much money is incoming in future. And similarly, it can tell how much money would be outgoing
    • For both the above, validation needs to be demonstrated through the models performance on the test data (assuming parts of csv file is split into train vs test data)
  • Implementation (a python script): Provided a transaction file, we should be able to forecast the cash balance and cash flow
Predict the next transaction:
  • Input for Learning/Training: Transaction set in a csv file. Relevant parameters includes the Date, Transaction details which is a text, Payer/Payee, Amount of Debit/Credit and Balance
  • Output from Learning/Training:  A model that can forecast the following:
    • Transactions Prediction – at least with some reliability for the next week/month and up to 6 future months. The model should also tell amount, the payer/payee and the possible future date of the transaction
    • For both the above, validation needs to be demonstrated through the models performance on the test data (assuming parts of csv file is split into train vs test data)
  • Implementation ( a python script): Provided a transaction file, we should be able to predict the future transactions

Leave Comment

Your email address will not be published. Required fields are marked *