Use cases & endpoints
Fed SEP projections data — growth, inflation, rates
The dot plot gets the attention, but the Summary of Economic Projections is a broader document: participants also submit projections for growth, unemployment and inflation across several horizons, published as central tendencies and full ranges. The relationship between those projections and the rate path is where the committee's reaction function becomes visible.
As a document it is readable and as data it is awkward — a set of tables whose structure has evolved, published a few times a year, with values that only make sense next to the round they came from.
Hindcast structures each round with every variable, horizon, central tendency and range preserved, and keeps every vintage. That makes the interesting comparisons mechanical: how a round revised its own prior projections, whether the rate path moved with the inflation projection or against it, and how far the range widened when the committee disagreed.
hindcast — session
previewhindcast> projections for growth and inflation from one SEP round
→ tool: cb_projections { bank:"FED", release:"...", variables:["gdp","pce"] }
✓ central tendency and range by horizon for that round
source: FOMC Summary of Economic Projections
What's covered
- All SEP variables, not only the policy rate projections
- Central tendencies and full ranges retained separately
- Projections by horizon including longer-run values
- Every round kept as a vintage with its release date
- Round-to-round revisions computable without manual alignment
- Linked to the statement, minutes and dot plot from the same meeting
Join the waitlist — this page's dataset is already selected.
Questions
What is the difference between the central tendency and the range?
The range covers all participants' projections; the central tendency excludes the highest and lowest few. The gap between them is a rough read on disagreement, which is why both are kept rather than collapsed into one summary number.
Are projections comparable across rounds?
Mostly, and the vintages make the exceptions visible. Definitions and horizons have shifted occasionally, so keeping each round intact is what allows a comparison to be checked rather than assumed.
Related
Join the waitlist. Vote the roadmap. First in, first served.