Model MatchHelp

Recipes

Summarize a market

One request for a metro's loan volume, units, average rate, loan size, credit score, LTV, and DTI, with the previous period for comparison.

Market analytics run over de-identified origination data, so they’re free of per-row credits. Name a geography and a period, and you get the current window and the one before it.

The request

curl -X POST https://api.modelmatch.com/v1/market/analytics/summary \
  -H "x-api-key: mm_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "flatFilters": { "metro": "12420" },
    "period": "last12Months"
  }'

With the MCP server connected, ask:

Summarize the Austin mortgage market over the last 12 months versus the year before: volume, units, average rate, loan size, and credit score.

metro is a 5-digit CBSA code. state, county, and zip work the same way. Market endpoints also accept rolling periods such as last90Days and previousMonth.

What comes back

{
  "current": {
    "originatedUnits": 19488,
    "originatedVolume": 7834183177.74,
    "units": 43477,
    "volume": 17410004985.1,
    "avgRate": 6.15,
    "avgLoanAmount": 400672,
    "avgCreditScore": 729,
    "avgLTV": 75.9,
    "avgDTI": 42.1,
    "avgIncome": 197411,
    "avgPurchasePrice": 490780,
    "outcomes": [
      { "outcome": "loan originated", "units": 19488, "volume": 7834183177.74 },
      { "outcome": "application withdrawn", "units": 6604, "volume": 2717032263.3 },
      { "outcome": "application denied", "units": 3322, "volume": 1111150760.7 }
    ]
  },
  "previous": {
    "originatedUnits": 22142,
    "originatedVolume": 8777604668.71,
    "avgRate": 6.4,
    "avgLoanAmount": 391169,
    "avgCreditScore": 731
  },
  "completeness": { "through": "2026-10-02" }
}
  • units and volume count every application, including denials and withdrawals. originatedUnits and originatedVolume are closed loans. Use the originated figures for market size.
  • outcomes partitions the applications, so you can compute denial and withdrawal rates directly.
  • completeness.through is the last date the data is complete to. The current window may be a few weeks behind today.

Make it yours

  • Purchase only: add "loanPurpose": "purchase" to flatFilters. loanType takes conventional, fha, va, or usda.
  • Monthly trend: call /v1/market/analytics/time-series with "interval": "monthly" and a measure. Add a forecast to project it forward.
  • Break it down: call /v1/market/analytics/chart with "slice": "loanType" or "slice": "county" for one bucket per value.
Try it Open in API reference Ask an assistant: marketAnalytics