Time Series • Introduction to Time Series

P1.06 Uses of Forecasting

Practical 1 Introduction to Time Series

Quick concept

Using last year's monthly demand, explain one business decision that a simple next-month forecast could support, and one decision it should not support.

Practical / Solution

P1.06 Uses of Forecasting

Problem Statement

Using last year's monthly demand, explain one business decision that a simple next-month forecast could support, and one decision it should not support.

Learning Outcomes

Connect a forecast to a practical use.
State that a forecast is an estimate, not a guarantee.

Hint

Inventory and staffing are typical uses. A 12-month investment plan needs more than one naive next-month number.

Theory

Forecasting estimates future values from historical information and a method. Uses include business demand, inventory, production, finance, energy and public planning.

Dataset / Data Source

Constructed monthly demand: 12 values from last year. Teaching data.

Analysis / Program

import pandas as pd idx = pd.date_range("2025-01-01", periods=12, freq="MS") demand = [90, 92, 95, 94, 100, 108, 115, 112, 105, 98, 96, 101] ts = pd.Series(demand, index=idx, name="demand") # Naive next-month idea: last observed value naive_next = ts.iloc[-1] print("Last observed month:", ts.index[-1].date()) print("Naive next-month estimate:", naive_next) print("Recent average (3 months):", round(ts.tail(3).mean(), 1))

Expected Output

Printed last date, naive estimate equal to the last value, and a 3-month average. These are calculated from the listed 12 numbers.

Result / Interpretation

The naive estimate can help a shop plan next month's stock. It should not be treated as a certain annual budget. Forecasting supports decisions; it does not replace judgement.

Note

A forecast is an estimate, not a guarantee.