Global price of Iron Ore (PIORECRUSDA) | FRED | St. Louis Fed
- Rows
- 34
- Columns
- 2
- Data range
- 1992-01-01 to 2025-01-01
Graph and download economic data for Global price of Iron Ore (PIORECRUSDA) from 1992 to 2025 about ore, iron, World, and price.
AI analysis
Dataset Analysis: Global Price of Iron Ore (PIORECRUSDA)
1. Dataset Overview and Value
This dataset tracks the annual global price of iron ore in USD from 1992 to 2025, sourced directly from the Federal Reserve Bank of St. Louis's FRED database (fredgraph.csv?id=PIORECRUSDA). As a Fed-maintained series, this data benefits from institutional credibility, standardized methodology, and regular update cycles—FRED typically refreshes commodity series as new World Bank or IMF source data becomes available. With 34 years of annual observations, this dataset is well-suited for long-run macroeconomic correlation studies, particularly around industrial demand cycles, steel production trends, infrastructure spending, and broader commodity supercycles. Its long time horizon (33+ years) makes it valuable for identifying structural breaks (e.g., China's industrialization post-2000) and testing lead-lag relationships with other economic indicators.
2. Data Quality Observations
Data quality appears strong at first glance: zero null cells across both columns and 34 distinct values in each, suggesting no missing observations or repeated/stale entries. However, the duplicate-row count is marked as pending recomputation ("Phase D — recomputed against Parquet"), so true uniqueness at the row level hasn't been fully verified yet—this should be confirmed before treating the series as gap-free. Additionally, the sample data rows are empty (0 of 34 shown), meaning no actual value inspection has occurred yet; formatting issues, unit inconsistencies, or placeholder values (e.g., zeros) can't be ruled out until Phase D wiring is complete. I'd flag this as "presumed clean, pending verification" rather than fully validated.
3. Key Column Distributions
The core metric, PIORECRUSDA, shows substantial volatility: it ranges from a low of $11.45 to a high of $167.79, with a mean of $63.45 and median of $57.35. The mean-median gap combined with a skewness of 0.515 indicates a moderate right skew—consistent with commodity price behavior, where prices spend long periods at lower baseline levels punctuated by sharp price spikes (e.g., the 2008 and 2021 commodity booms). The interquartile range (Q1 = $13.13, Q3 = $106.98) is unusually wide relative to the median, reflecting the boom-bust nature of iron ore pricing rather than a normally distributed series. This wide spread signals that raw-level correlation analysis may benefit from log-transformation or year-over-year percentage change calculations to stabilize variance before regression or correlation testing.
4. Recommended Join Keys
The observationdate column (Date type, 0 nulls, 34 distinct values) is the clear and only viable join key. Since the data is annual, any cross-dataset correlation work should either (a) join to other annual series on year, or (b) if joining to monthly/quarterly data, aggregate the higher-frequency series to annual averages or year-end snapshots first to avoid misalignment. Given the small n (34 rows), care should be taken with degrees of freedom in any multi-variable regression.
5. Complementary Datasets for Correlation Discovery
Strong candidates for pairing include: steel production indices (e.g., World Steel Association output data) and China's industrial production/GDP (given China consumes ~70% of seaborne iron ore), other FRED commodity series like coal, copper, or the broader PALLFNFINDEXA (all commodities index) for supercycle comparison, Baltic Dry Index (shipping/freight rates, since iron ore is bulk-shipped), exchange rate series (especially AUD/USD and BRL/USD, since Australia and Brazil are top exporters), and inflation/PPI indices to distinguish real vs. nominal price movements. Given the annual frequency, this dataset pairs best with other annual macro series rather than high-frequency financial data.
Columns
- observation_date (date)
- PIORECRUSDA (decimal)