Correlation

Compute pairwise Pearson correlation coefficients for any combination of crypto assets and traditional finance instruments. The resulting NxN matrix is built from daily log returns and supports lookback periods from 30 days to 3 years. Use it for portfolio diversification analysis, regime detection, or cross-asset risk monitoring.

Get correlation matrix

Returns a symmetric correlation matrix for the requested assets and period. The matrix is computed from daily log returns with a minimum data coverage threshold of 80% -- asset pairs with insufficient overlapping data return null coefficients.

Query parameters

Name
period
Type
string
Description
Lookback period. One of 30d, 90d, 1y, 3y. Defaults to 30d.
Name
ids
Type
string
Description
Comma-separated asset identifiers. Crypto assets use canonical IDs (e.g., bitcoin, ethereum, solana). TradFi assets use short IDs (e.g., sp500, gold, dxy, aapl, nvda). Maximum 10 assets. Defaults to bitcoin,ethereum,solana,sp500,gold.

Response fields

Name
assets
Type
array
Description
Ordered array of asset metadata matching the matrix dimensions.
Name
assets[].id
Type
string
Description
Asset identifier (canonical crypto ID or TradFi short ID).
Name
assets[].name
Type
string
Description
Display name (e.g., Bitcoin, S&P 500).
Name
assets[].symbol
Type
string
Description
Ticker symbol (e.g., BTC, SPY).
Name
assets[].type
Type
string
Description
"crypto" or "tradfi".
Name
assets[].logoUrl
Type
string
Description
URL to the asset logo, if available.
Name
matrix
Type
(number | null)[][]
Description
NxN array of Pearson correlation coefficients. matrix[i][j] is the correlation between assets[i] and assets[j]. Values are in [-1, 1], or null when fewer than 80% of the period's days are present in both series. Self-correlations on the diagonal are 1.
Name
dataPoints
Type
number[][]
Description
NxN array of integers. dataPoints[i][j] is the number of overlapping daily observations used to compute matrix[i][j]. TradFi-only pairs use trading-day counts (≈252/year), so a 90-day period yields fewer points than a crypto-only pair over the same window.
Name
period
Type
string
Description
The period that was applied.
Name
computedAt
Type
string
Description
ISO 8601 timestamp of the last matrix computation, or null.

Request

GET

/v1/correlation/matrix

curl -G https://www.sharpe.ai/api/v1/correlation/matrix \
  -H "Authorization: Bearer sk_live_your_key_here" \
  -d period=90d \
  -d ids=bitcoin,ethereum,solana,sp500,gold

{
  "data": {
    "assets": [
      {
        "id": "bitcoin",
        "name": "Bitcoin",
        "symbol": "BTC",
        "type": "crypto",
        "logoUrl": "https://assets.sharpe.ai/coins/bitcoin.png"
      },
      {
        "id": "ethereum",
        "name": "Ethereum",
        "symbol": "ETH",
        "type": "crypto",
        "logoUrl": "https://assets.sharpe.ai/coins/ethereum.png"
      },
      {
        "id": "solana",
        "name": "Solana",
        "symbol": "SOL",
        "type": "crypto",
        "logoUrl": "https://assets.sharpe.ai/coins/solana.png"
      },
      {
        "id": "sp500",
        "name": "S&P 500",
        "symbol": "SPY",
        "type": "tradfi",
        "logoUrl": "/tradfi-logos/spy.svg"
      },
      {
        "id": "gold",
        "name": "Gold",
        "symbol": "GLD",
        "type": "tradfi",
        "logoUrl": "/tradfi-logos/gld.svg"
      }
    ],
    "matrix": [
      [1.0,   0.87,  0.82,  0.45,  0.12],
      [0.87,  1.0,   0.91,  0.41, -0.03],
      [0.82,  0.91,  1.0,   0.38, -0.08],
      [0.45,  0.41,  0.38,  1.0,   0.15],
      [0.12, -0.03, -0.08,  0.15,  1.0]
    ],
    "dataPoints": [
      [90, 90, 90, 64, 64],
      [90, 90, 90, 64, 64],
      [90, 90, 90, 64, 64],
      [64, 64, 64, 64, 64],
      [64, 64, 64, 64, 64]
    ],
    "period": "90d",
    "computedAt": "2026-03-28T04:00:00Z"
  },
  "meta": {
    "request_id": "req_abc123def456ghij",
    "timestamp": "2026-03-28T12:00:00Z"
  }
}

Available TradFi assets

The following TradFi asset IDs are supported in the ids parameter. Crypto assets use their canonical ID (e.g., bitcoin, ethereum, solana, dogecoin).

Indices and commodities

ID	Name	Symbol
`sp500`	S&P 500	SPY
`nasdaq`	Nasdaq 100	QQQ
`gold`	Gold	GLD
`dxy`	US Dollar Index	DXY

US stocks

ID	Name	Symbol
`aapl`	Apple	AAPL
`msft`	Microsoft	MSFT
`nvda`	NVIDIA	NVDA
`tsla`	Tesla	TSLA
`amzn`	Amazon	AMZN
`googl`	Alphabet	GOOGL
`meta`	Meta	META
`brk`	Berkshire	BRK.B
`jpm`	JPMorgan	JPM
`v`	Visa	V

Additional stocks are available on demand: amd, nflx, crm, dis, ba, ko, pep, wmt, cost, intc, pypl, coin, mstr, pltr, uber, sq, snow, sofi, hood, spot.

Request — mixed assets

GET

/v1/correlation/matrix?ids=bitcoin,nvda,sp500

curl -G https://www.sharpe.ai/api/v1/correlation/matrix \
  -H "Authorization: Bearer sk_live_your_key_here" \
  -d period=1y \
  -d ids=bitcoin,nvda,sp500

{
  "data": {
    "assets": [
      {
        "id": "bitcoin",
        "name": "Bitcoin",
        "symbol": "BTC",
        "type": "crypto"
      },
      {
        "id": "nvda",
        "name": "NVIDIA",
        "symbol": "NVDA",
        "type": "tradfi",
        "logoUrl": "/tradfi-logos/nvda.svg"
      },
      {
        "id": "sp500",
        "name": "S&P 500",
        "symbol": "SPY",
        "type": "tradfi",
        "logoUrl": "/tradfi-logos/spy.svg"
      }
    ],
    "matrix": [
      [1.0,  0.52, 0.48],
      [0.52, 1.0,  0.89],
      [0.48, 0.89, 1.0]
    ],
    "dataPoints": [
      [365, 252, 252],
      [252, 252, 252],
      [252, 252, 252]
    ],
    "period": "1y",
    "computedAt": "2026-03-28T04:00:00Z"
  },
  "meta": {
    "request_id": "req_def456ghi789jklm",
    "timestamp": "2026-03-28T12:00:00Z"
  }
}

Interpreting correlation values

Range	Interpretation
0.8 to 1.0	Strong positive correlation -- assets move together
0.5 to 0.8	Moderate positive correlation
0.0 to 0.5	Weak positive correlation
-0.5 to 0.0	Weak negative correlation
-0.8 to -0.5	Moderate negative correlation
-1.0 to -0.8	Strong negative correlation -- assets move opposite

TradFi assets (stocks, indices) only have data on trading days, so dataPoints will be lower than crypto pairs for the same period. A 90-day period yields ~64 TradFi data points versus 90 for crypto-only pairs.

Read the error reference