Industrial Carbon Dioxide Emissions, Petroleum for United (DISCONTINUED) (EMISSCO2TOTVICPEA) | FRED | St. Louis Fed
- Rows
- 38
- Columns
- 2
- Data range
- 1980-01-01 to 2017-01-01
Graph and download economic data for Industrial Carbon Dioxide Emissions, Petroleum for United (DISCONTINUED) (EMISSCO2TOTVICPEA) from 1980 to 2017 about petroleum, carbon dioxide emissions, industry, and USA.
AI analysis
Dataset Analysis: Industrial CO2 Emissions from Petroleum (EMISSCO2TOTVICPEA)
1. Dataset Overview and Value
This dataset, sourced from the St. Louis Fed's FRED database (fredgraph.csv API), tracks annual industrial carbon dioxide emissions from petroleum consumption in the United States from 1980 to 2017. Notably, the series is discontinued, meaning no new observations will be added beyond 2017 — an important caveat for any longitudinal or forward-looking analysis. As a FRED-hosted series, the data benefits from a credible, well-documented government/institutional pipeline, making it a reliable historical reference point for energy-economy research. Its value lies primarily in retrospective correlation studies examining relationships between industrial fossil fuel emissions, economic activity (GDP, industrial production indices), energy prices, and policy interventions (e.g., EPA regulations, Kyoto Protocol-era commitments) during the 1980–2017 window.
2. Data Quality Observations
Data quality appears strong on the surface: zero null cells across both columns and all 38 rows, with each column showing 38 distinct values (i.e., no repeated dates or emission values), suggesting no obvious duplication within the ingested set. However, duplicate-row verification is marked as pending ("Phase D — recomputed against Parquet"), so full deduplication confidence is not yet established. The dataset is also small (only 38 annual observations), which limits statistical power for regression or time-series modeling and makes it more suited to trend/correlation analysis than robust econometric inference. Sample row data is unavailable in this profile, limiting immediate spot-checking of raw values.
3. Key Column Distributions
The core metric, EMISSCO2TOTVICPEA, ranges from 325.47 to 499.56 (likely million metric tons CO2), with a mean of 378.92 and median of 373.19 — the mean exceeding the median indicates a right-skewed distribution, confirmed by a skewness value of 1.69. This suggests a small number of high-emission years pulling the distribution upward, consistent with the single detected outlier. The interquartile range (Q1=362.05, Q3=390.10) is fairly tight relative to the full range, implying most annual values cluster in the mid-300s-to-390 band, with the outlier (likely a peak year, possibly late-1990s/2000s pre-recession industrial activity) driving the elevated max and skew. The standard deviation of ~30.49 relative to the mean indicates moderate year-over-year variability (~8% CV).
4. Recommended Join Keys
The observationdate column (annual granularity, 1980–2017) is the natural and only viable join key. For cross-dataset correlation, analysts should align other series to annual periodicity — either by using year-only extraction (stripping month/day) or by resampling higher-frequency datasets (monthly/quarterly) to annual averages/sums before joining on date.
5. Complementary Datasets for Correlation Analysis
Strong pairing candidates include: U.S. GDP and Industrial Production Index (INDPRO) to test emissions-output decoupling; petroleum consumption/price series (e.g., WTI crude prices, EIA petroleum consumption data) to explore price-elasticity of emissions; other FRED CO2 emissions series (coal, natural gas, or total emissions) for fuel-mix substitution analysis; EPA/EIA energy intensity metrics; and macroeconomic indicators like unemployment rate or manufacturing sector value-added, which could reveal how economic cycles (e.g., 2008 recession) correlate with industrial emissions declines.
Columns
- observation_date (date)
- EMISSCO2TOTVICPEA (decimal)