Natural Gas Prices (Henry Hub)

Rows: 7,379
Columns: 2

Daily natural gas spot prices at the Henry Hub in Louisiana, a key benchmark for the North American energy market.

AI analysis

Natural Gas Prices (Henry Hub) — Dataset Analysis

Dataset Overview & Value This dataset captures daily natural gas spot prices at the Henry Hub in Louisiana, the primary pricing benchmark for North American natural gas markets. Sourced from [datahub.io/core/natural-gas](https://datahub.io/core/natural-gas/r/daily.csv) — a well-maintained open data registry — the dataset spans 7,379 trading days, suggesting roughly 28–30 years of daily coverage (likely early 1990s through the early 2020s), making it exceptionally well-suited for long-horizon time-series and correlation studies. Its value lies in being a foundational energy price series that ripples through electricity markets, industrial costs, consumer utility bills, and broader macroeconomic conditions, offering rich potential for cross-sector correlation discovery.

Data Quality Data quality here is remarkably strong. There is only 1 null cell across the entire dataset, confined to the Price column (0.01% null rate), which is negligible and unlikely to affect any analysis. The duplicate row count is noted as pending Phase D recomputation, but given that Date has exactly 7,379 distinct values matching the total row count perfectly, there are effectively zero duplicate rows — every record represents a unique trading date. There are no type mismatches reported, and both columns appear clean and well-typed. The one missing price value could be trivially imputed (e.g., linear interpolation) or simply dropped without consequence.

Key Column Distributions The Price column is where the analytical richness lives. With a mean of \$4.08 and a median of \$3.36, the distribution is right-skewed (skewness = 2), meaning a relatively small number of high-price episodes pull the average upward — a classic pattern for commodity prices driven by supply shocks. The interquartile range (Q1 = \$2.61, Q3 = \$5.03) confirms that the bulk of prices cluster in a moderate range, while the full spread from \$1.05 to \$30.72 reflects extreme volatility events (e.g., Winter Storm Uri in 2021 or the 2005 post-hurricane spikes). The 286 outliers (~3.9% of records) deserve careful treatment — they likely correspond to real market events rather than data errors, and should be preserved but flagged in any modeling work.

Recommended Join Key The Date column is the clear and only join key, and it is ideal for this purpose: zero nulls, 7,379 distinct values, perfect uniqueness. It should be treated as a daily time-series index. When joining to other datasets, be mindful of trading-day vs. calendar-day alignment — some datasets report on weekdays only, while others (weather, for example) report every calendar day. A date-normalized join with forward-fill logic is recommended for handling gaps across datasets.

Suggested Correlation Datasets Several dataset categories would pair powerfully with this data. Electricity prices (e.g., EIA wholesale power prices by region) are a natural first choice, since natural gas is a dominant fuel for electricity generation and price pass-through is well-documented. Crude oil and petroleum prices (WTI or Brent) would reveal energy commodity co-movement and divergence periods. U.S. weather and temperature data (heating/cooling degree days) would help explain seasonal demand spikes and could surface strong winter-price correlations. Industrial production indices or GDP growth data from FRED would test macroeconomic sensitivity. Finally, LNG export volumes or pipeline flow data post-2016 would be particularly interesting to examine whether export capacity growth has structurally altered domestic price dynamics.

Columns

Date (date)
Price (decimal)