Python Essentials for Finance in One Session: Data Types, Lists, Dictionaries, Loops, Functions and Error Messages

What you will learn

• The data types you will meet in price data: int, float, str, bool, None and datetime.
• How lists and dictionaries store a trade list, and how a for loop processes it.
• How to write a function with a docstring and a default argument, and why functions make backtests testable.
• How to read a Python traceback from the bottom up and fix the five most common errors.

Explanation

Types. Every value in Python has a type. 42 is an int (whole number), 42.5 a float (decimal number stored in binary), "AAPL" a str (text), True a bool, and None means "no value". Floats are approximate (0.1 + 0.2 prints 0.30000000000000004), so never compare them with ==; use round() or a tolerance. For dates use pandas timestamps (lesson 3).

Lists are ordered collections written with square brackets: closes = [100.0, 102.0, 101.5]. Indexing starts at zero (closes[0] is 100.0), negative indices count from the end (closes[-1] is the latest close), and slicing closes[1:3] returns elements 1 and 2 but not 3. The same end-excluded rule applies to pandas iloc.

Dictionaries map keys to values: trade = {"symbol": "AAA", "entry": 50.0, "qty": 100}. Access with trade["entry"]; a missing key raises KeyError, while trade.get("fee", 0.0) returns a default.

Loops and conditions. A for loop visits each element: for t in trades: followed by an indented block. Indentation (four spaces) defines the block. if, elif and else branch on conditions, and comparison operators are ==, !=, <, >=.

Functions package logic so it can be reused and tested. The keyword def starts a definition, parameters go in brackets, and return sends a value back:

def r_multiple(entry, exit_price, stop, side="long"):

'''Return profit in units of initial risk (R).'''

risk = abs(entry - stop)

pnl = exit_price - entry if side == "long" else entry - exit_price

return pnl / risk

The triple-quoted string is a docstring (documentation), and side="long" is a default argument. Keep functions small, explicit about their inputs and free of side effects; that is what lets lesson 9 unit-test them.

Reading errors. When something fails, Python prints a traceback. Read the last line first: it names the error type and message. Then look at the line just above it, which shows where in your code it happened. The most common errors: NameError (you used a variable before defining it, often because a notebook cell was not run); TypeError (for example adding a string from a CSV file to a number, fix with float()); KeyError (a column or dictionary key is spelled differently, check for spaces and capital letters); IndexError (you asked for element 10 of a list with 10 elements, remember zero-based indexing); and ZeroDivisionError (a stop equal to the entry gives risk zero).

Worked example

Hypothetical example: three closed long trades, all in euros, commissions and taxes ignored.

trades = [

{"symbol": "AAA", "entry": 50.0, "exit": 53.0, "stop": 48.0, "qty": 100},

{"symbol": "BBB", "entry": 20.0, "exit": 19.0, "stop": 19.0, "qty": 250},

{"symbol": "CCC", "entry": 80.0, "exit": 78.0, "stop": 76.0, "qty": 60},

]

total = 0.0

for t in trades:

pnl = (t["exit"] - t["entry"]) * t["qty"]

r = r_multiple(t["entry"], t["exit"], t["stop"])

total += pnl

print(t["symbol"], pnl, round(r, 2))

print("Total:", total)

AAA profit = (53 − 50) × 100 = €300; risk per share €2, so R = 3 ÷ 2 = 1.5. BBB = (19 − 20) × 250 = −€250; risk €1 per share, R = −1.0 (stopped out exactly). CCC = (78 − 80) × 60 = −€120; risk €4, R = −0.5. Total = 300 − 250 − 120 = −€70. Output lines: AAA 300.0 1.5, BBB -250.0 -1.0, CCC -120.0 -0.5, Total: -70.0.

total += pnl is shorthand for total = total + pnl. Three trades say nothing about a method; the point is the mechanics.

Common mistake

Practice activity

On paper, write a function position_size(account, risk_pct, entry, stop) that returns the whole number of shares (rounding down) and raises a clear ValueError if the stop equals the entry. Then write the result for account €8,000, risk 1.25%, entry €32.50, stop €30.00. Finally, name the error type Python would raise if you called the function with entry="32.50" (a string) and did not convert it.

Key takeaways

• Floats are approximate, indexing starts at zero and slices exclude the end; these three facts prevent most beginner bugs.
• A list of dictionaries is the natural shape for trades before pandas; loops process it, functions give it reusable logic.
• Write small, pure functions with docstrings; they are the units you will test in lesson 9.
• Read tracebacks from the bottom up, and restart-and-run-all before trusting a notebook result.

Next step

Lesson 3 introduces pandas: Series, DataFrames, the DatetimeIndex, selecting and filtering rows, and turning daily bars into weekly bars with resample.