Transaction DataFrame Standard

A Python library for standardizing transaction data from multiple financial sources into a unified format for net wealth analysis and expense tracking in Pandas.

Overview

This library provides:

1. Comprehensive Standard Schema - A unified 17-column format supporting both cash and investment transactions
2. Adapters - Transform data from various sources (banks, investment platforms, budgeting apps) into the standard format
3. Analysis Functions - Process standardized data for insights (expense analysis, net wealth tracking, savings rates)

Getting Started

The best way to get started is to take a look at the tests folder, which provide up to date examples on how to use this library.

The Comprehensive Standard

The library uses a comprehensive standard that supports:

• Cash transactions: Bank accounts, credit cards, budgeting apps (Monzo, YNAB, Bank of America)
• Investment transactions: Stocks, funds, ETFs (Vanguard, Hargreaves Lansdown, Halifax)
• Pensions: SIPP, Personal Pension, Workplace Pension
• Company accounts: Limited company transactions

Standard Columns (17 total)

Column	Type	Description	Example
date	datetime	Transaction date	2024-01-15
time	string	Transaction time (optional)	14:23:45
account	string	Account identifier	"Monzo", "Vanguard ISA"
account_type	string	Account classification	"Current", "ISA", "SIPP"
transaction_type	string	Transaction category	"Expense", "Buy", "Dividend"
category	string	User-defined category	"Groceries", "Index Funds"
amount	float	Transaction amount	-45.67, 500.00
currency	string	Currency code	"GBP", "USD"
asset_ticker	string	Investment ticker (optional)	"VWRL", "META", None
units	float	Shares/units (optional)	6.234, None
price_per_unit	float	Price per share (optional)	80.21, None
notes	string	Additional description	"Tesco - weekly shopping"
country	string	Transaction country (optional)	"UK", "Thailand"
city	string	Transaction city (optional)	"London", "Bangkok"
is_pension_contribution	bool	Pension contribution flag	True, False
data_source	string	Source file/system	"Monzo_Export_2024.csv"
data_quality	string	Data reliability	"Verified", "Estimated"

Standard Enums

TransactionType: Income, Expense, Transfer, Buy, Sell, Dividend, Interest, Fee, Contribution, Withdrawal, Liquidation

AccountType: Current, Savings, Credit Card, ISA, SIPP, Personal Pension, Workplace Pension, General Investment, Ltd, Help to Buy ISA

DataQuality: Verified, Estimated, Reconstructed

See src/transaction_dataframe_standard/standard.py for full documentation and examples.

Adapters

Adapters transform source data into the standard format:

Available Adapters

• MonzoTransactionsAdapter - Monzo bank CSV exports
• BOATransactionsAdapter - Bank of America credit card statements (partial)
• HLTransactionsAdapter - Hargreaves Lansdown quarterly PDFs (in development)
• VanguardTransactionsAdapter - Vanguard Excel exports (in development)
• YNABTransactionsAdapter - YNAB budget register CSV (in development)

Example Usage

from transaction_dataframe_standard.adapters import MonzoTransactionsAdapter
from transaction_dataframe_standard.standard import validate_standard_dataframe

# Load and transform Monzo data
adapter = MonzoTransactionsAdapter('path/to/monzo_export.csv')
df = adapter.transactions

# Validate the output
is_valid, errors = validate_standard_dataframe(df)
if not is_valid:
    print(f"Validation errors: {errors}")

Analysis Functions

Process standardized data for insights. See src/transaction_dataframe_standard/functions/ for available functions.

Example: Expense Analysis

from transaction_dataframe_standard.functions import print_expense_stats_trailing_twelve_months

# Analyze expenses over rolling 12-month periods
print_expense_stats_trailing_twelve_months(
    df,
    exclude_names=['Vanguard', 'HL']  # Exclude investments
)

Legacy Format

The original 8-column format for cash-only transactions is still supported in existing Monzo/BOA adapters: Date, Time, Type, Name, Category, Description, Money Out, Money In

New adapters should use the comprehensive 17-column standard.