Coinbase Bitcoin (CBBTCUSD) | FRED | St. Louis Fed
- Rows
- 4,230
- Columns
- 2
- Data range
- 2014-12-01 to 2026-06-30
Graph and download economic data for Coinbase Bitcoin (CBBTCUSD) from 2014-12-01 to 2026-06-30 about cryptocurrency and USA.
AI analysis
Data Quality & Structural Analysis: Coinbase Bitcoin (CBBTCUSD)
1. Dataset Overview & Value
This dataset tracks the daily USD price of Bitcoin as sourced from Coinbase, distributed through FRED (Federal Reserve Economic Data), a highly credible publisher maintained by the St. Louis Fed. The source URL (fredgraph.csv?id=CBBTCUSD) confirms this is pulled via FRED's live API, meaning it reflects the Fed's official ingestion of Coinbase exchange data rather than a static or scraped snapshot — this lends strong institutional credibility to timestamps and values. Notably, the stated coverage window extends to 2026-06-30, which is likely a placeholder/forward-dated series boundary rather than actual future observations (worth verifying against the null count concentration). Because FRED cross-publishes thousands of macroeconomic series on the same daily date index, this dataset is exceptionally well-positioned for correlation studies linking crypto-asset pricing to traditional financial and economic indicators.
2. Data Quality Observations
The dataset is structurally clean at the row level — 4,230 rows with 4,230 distinct observationdate values and zero nulls in the join key, indicating no duplicate or missing dates. The only quality concern is concentrated entirely in the CBBTCUSD column, which carries all 35 null cells (0.8% of its values, and 100% of total null cells in the dataset). This is a modest but non-trivial gap — likely explained by exchange holidays, data feed interruptions, or the forward-dated portion of the series (if the 2026 endpoint includes not-yet-realized dates). Duplicate-row computation is still pending (marked for Phase D/Parquet recheck), so full deduplication confidence isn't yet available, but the distinct-count parity on the date column is a good preliminary signal of row-level uniqueness.
3. Key Column Distributions
The CBBTCUSD column shows classic cryptocurrency price behavior: a wide range (min=120, max=124,720) with a mean (29,580.5) well above the median (14,160), producing a positive skew of 1.12 — consistent with Bitcoin's historical pattern of long consolidation periods punctuated by explosive rallies. The interquartile range (Q1=3,879.55, Q3=48,162.6) shows that most historical observations cluster at much lower price levels than the current market has reached, reflecting Bitcoin's multi-year growth trajectory. The standard deviation (σ=32,561.9) nearly matches the mean, underscoring extreme volatility. 60 statistical outliers are flagged, almost certainly clustered in Bitcoin's parabolic peak periods (e.g., late 2017, 2021, and more recent bull cycles) rather than data errors — analysts should verify these aren't ingestion artifacts before treating them as genuine market extremes.
4. Recommended Join Key
observationdate is the clear and only sensible join key — it's complete (zero nulls), fully distinct, and daily-grained, matching FRED's standard convention across nearly all its economic series. Analysts should align on daily or resample to weekly/monthly frequency (using close-of-period or averaging) when joining against series with different native frequencies (e.g., monthly CPI or quarterly GDP).
5. Suggested Pairings for Correlation Analysis
Strong candidates include: (a) DXY / broad USD index and other FRED FX series to test Bitcoin's "digital gold" or dollar-hedge narrative; (b) DGS10 (10-Year Treasury Yield) and Federal Funds Rate (FEDFUNDS) to explore risk-on/risk-off and monetary policy sensitivity; (c) CPI (CPIAUCSL) for inflation-hedge testing; (d) S&P 500 / NASDAQ equity indices to assess Bitcoin's correlation with risk assets versus its "uncorrelated asset" reputation; (e) VIX for volatility regime comparisons; and (f) M2 Money Supply (M2SL) to examine liquidity-driven price movements. Given Bitcoin's daily granularity and long time span, this dataset is also suitable for lead-lag and rolling-correlation analyses against these macro series to detect regime shifts (e.g., pre/post-2020 monetary expansion, rate-hike cycles).
Columns
- observation_date (date)
- CBBTCUSD (decimal)