US recession-probability signals (NY Fed smoothed probabilities)
Monthly US recession-probability signals from the New York Fed's Smoothed U.S. Recession Probabilities (Chauvet-Piger dynamic-factor Markov-switching model, redistributed by FRED, 1967 ->): the forward-looking probability that the economy is in recession, with 3-month momentum, 30-month change volatility, 3-sigma anomaly flags, drift forecasts, a 5-year probability z-score, elevated (>=30) and recession-call (>=50) flags, a surging-risk flag, and a high-probability streak counter. The forward-looking companion to the NBER-based realized recession flag in us-output-business-cycle-signals. All rows normalized to country_code USA. Raw series: Federal Reserve Bank of New York via FRED.
Quality
Attribution
Federal Reserve Bank of New York (Smoothed U.S. Recession Probabilities) via FRED; derived signals by Frontier Data Hub
Schema
| Column | Type | Description |
|---|---|---|
| date | string | Observation date (FRED API field date; YYYY-MM-DD, monthly). |
| country | string | |
| country_code | string | |
| series_id | string | FRED series ID: RECPROUSM156N (Smoothed U.S. Recession Probabilities, Federal Reserve Bank of New York). |
| series_label | string | Official FRED series title as published for the series. |
| value | float | Smoothed probability that the U.S. economy was in a recession during the month, in percent, estimated from a dynamic-factor Markov-switching model of coincident indicators (Federal Reserve Bank of New York, via FRED). |
| momentum_3m | float | |
| volatility_30d | float | |
| anomaly_flag | integer | |
| forecast_1m | float | |
| prob_z_5y | float | |
| elevated_flag | integer | |
| recession_call_flag | integer | |
| rising_flag | integer | |
| high_prob_streak | integer |
Sample rows
| date | country | country_code | series_id | series_label | value | momentum_3m | volatility_30d | anomaly_flag | forecast_1m | prob_z_5y | elevated_flag | recession_call_flag | rising_flag | high_prob_streak |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1967-06-01 | United States | USA | RECPROUSM156N | Smoothed U.S. Recession Probabilities | 1.1 | — | — | 0 | — | — | 0 | 0 | 0 | 0 |
| 1967-07-01 | United States | USA | RECPROUSM156N | Smoothed U.S. Recession Probabilities | 0.54 | — | — | 0 | — | — | 0 | 0 | 0 | 0 |
| 1967-08-01 | United States | USA | RECPROUSM156N | Smoothed U.S. Recession Probabilities | 0.12 | — | — | 0 | — | — | 0 | 0 | 0 | 0 |
| 1967-09-01 | United States | USA | RECPROUSM156N | Smoothed U.S. Recession Probabilities | 0.52 | -0.5800000000000001 | — | 0 | — | — | 0 | 0 | 0 | 0 |
| 1967-10-01 | United States | USA | RECPROUSM156N | Smoothed U.S. Recession Probabilities | 0.18 | -0.36000000000000004 | — | 0 | — | — | 0 | 0 | 0 | 0 |
Download sample data
Download the full sample snapshot for this dataset (sample rows, not the complete dataset).
Use with an LLM
Point any LLM at the metadata endpoint — the documentation above is machine-readable too (JSON-LD + Croissant).
cURL
curl "https://datazimuts.com/v1/datasets/recession_prob_signals/us_recession_probability_signals" | jq '{title, rows, columns_count, license}'Python
import requests
ds = requests.get("https://datazimuts.com/v1/datasets/recession_prob_signals/us_recession_probability_signals").json()
print(ds["title"], ds["rows"], "rows")
# Sample rows for an LLM context window
for row in ds.get("sample_rows", [])[:5]:
print(row)API endpoint: https://datazimuts.com/v1/datasets/recession_prob_signals/us_recession_probability_signals
Tip: fetch /llms.txt for the full machine-readable catalog.