Direct answer
A worked example of Core CPI is a step-by-step numerical illustration of how “core inflation” can be calculated from an assumed set of CPI components when you exclude specific items (commonly volatile categories). The point of the example is not to predict anything, but to show the arithmetic and the assumptions that connect CPI component data to a Core CPI-style rate.
Core CPI: what the concept means (non-technical definition)
Core CPI refers to a version of the Consumer Price Index (CPI) that removes certain categories from the overall basket. The excluded categories are typically those with higher short-term volatility, so the remaining measure is intended to better reflect broader, underlying price pressures than the total CPI.
Two important things to separate are:
- Stable mechanics: how you compute an inflation rate from two CPI levels (or from component indexes).
- Variable provider details: which categories are excluded, what weights are used in the basket, how indexes are rebased, and how often data are revised.
Because those details can differ, a “Core CPI worked example” must state its own definitions and assumptions clearly.
Evidence or example: a transparent numerical worked scenario
Below is a fully specified example that uses the same inflation-rate logic you would apply to a published Core CPI series, but it uses hypothetical numbers.
Assumptions
- You have a core CPI index level (or an equivalent core-weighted index) for two time points.
- The index is constructed so that a percentage change over time represents inflation.
- No currency conversion is needed (CPI is typically a price-level index).
- “Core CPI month-over-month” uses the ratio of the core index between two consecutive months.
Step 1: Choose hypothetical core index levels
Assume:
- Core CPI index in Month A = 200.0
- Core CPI index in Month B = 203.0
Step 2: Compute the month-over-month core inflation rate
Inflation rate from A to B (in %) is:
- (203.0 / 200.0 − 1) × 100
- = (1.015 − 1) × 100
- = 1.5%
Step 3: Compute an annualized “year-over-year” core inflation rate (scenario)
Now assume:
- Core CPI index in Month A last year = 195.0
- Core CPI index in Month B this year = 203.0
Year-over-year core inflation is:
- (203.0 / 195.0 − 1) × 100
- = (1.041025… − 1) × 100
- ≈ 4.10%
Step 4: What this example does not assume
This scenario does not show how the core index is built from underlying components (such as which exact categories are excluded, or the detailed weighting). Instead, it demonstrates the calculation once you already have a “core CPI index level.” If you want to go one step deeper, you can treat the “core index” as the weighted result of included components—then you would also need explicit assumptions about exclusions and weights.
Limitations and risks (at least one failure mode)
-
Definition mismatch (failure mode): “Core CPI” may exclude different items depending on the country and the reporting methodology. If your worked example excludes categories that the published index does not, the computed rate won’t match reality.
-
Timing and revisions: CPI data can be updated or revised. If you compare a worked example’s month A/B with a later revised dataset, the numbers can differ even if the underlying computation method is correct.
-
Basket weights and re-rebasing: If the core index uses a basket with changing weights or a different base period, then component changes can translate into different index movements than your assumed index levels.
-
Index-level vs component-level confusion: Many readers try to rebuild a core index from components without the exact weights and exclusions. This can produce an internally consistent calculation that still diverges from the official core CPI series.
Verification and next question
To independently verify any Core CPI worked example, you should match the example’s assumptions to the published methodology:
- confirm what is excluded (the excluded categories),
- confirm the index type and base,
- confirm the timing convention (which month is compared to which),
- confirm whether you are using index levels (typical) or component price changes (different mechanics).