Datadory notebook

How Data Scientists Use Industrial REITs Data

Data scientists use industrial REITs data as a nearly-free modeling stack: SEC EDGAR's XBRL companyfacts API for fundamentals keyed by CIK, Nareit's T-Tracker and FTSE Nareit return files as sector targets and benchmarks, and Port of LA TEUs plus Census wholesale inventories as demand features. Thirteen of the 15 datasets Datadory catalogs for the industry are free.

1,744 datasets. Pick your catch.

How do you build a fundamental feature store from SEC XBRL?

Companyfacts is the highest-scoring record in this slice (quality 10) and the anchor of any entity panel: key-less JSON on data.sec.gov exposing every XBRL fact a registrant has ever reported. A minimal pipeline:

  1. Fix the universe. Pull Nasdaq's screener URL once - 7,079 securities across NYSE, NASDAQ and AMEX, 226 tagged Real Estate Investment Trusts, about 2.2 MB of JSON, no key - then intersect with StockAnalysis.com’s tighter 15-row REIT Industrial list to settle the narrow sector definition before you spend requests.
  2. Resolve symbols to CIKs and request the companyfacts document per registrant - about 3.5 MB of JSON for a large filer, holding 586 us-gaap concepts for Prologis alone.
  3. Select before you load. Filter us-gaap down to the statement lines you model; hundreds of concepts per filer are noise for a sector panel.
  4. Keep the vintages. Each fact carries a source filing reference, so restatements surface as extra dated values rather than overwrites - Prologis revenue facts run from CY2016 through CY2026Q1, and older filings stay attributable.
  5. Add the event stream. The submissions JSON behind EDGAR’s filing index lists roughly 1,000 recent filings per registrant plus paginated archives of about 2,000 more, with Prologis history verified back to September 18, 1997 - join it for announcement-date alignment in event studies.
  6. Respect fair access. Declare a User-Agent identifying your application, stay near SEC's documented 10 requests per second, and schedule off-peak for bulk refreshes.

One structural caveat belongs in your data dictionary: XBRL depth depends on when a filer started tagging, so factors that need pre-XBRL history require a different source than this API.

Which series can you actually backtest industrial REIT strategies against?

Nareit supplies both halves of the benchmark layer, and they play different roles. The FTSE Nareit US Real Estate Index Series ships 21 sector .xls files plus two index-level files refreshed monthly on reit.com, with annual returns beginning in 1972 and multi-decade monthly history per sector - the dedicated Industrial file is the natural total-return series for factor and correlation studies. The T-Tracker is the operations panel: about 274 labelled variables across 107 quarterly columns of FFO, NOI, same-store NOI, dividends, occupancy, leverage and cap rates for 21 equity and mortgage sectors, running from 2000 Q1 through 2026 Q2 as a 1.3 MB XLSX (alongside a 10.8 MB PDF commentary your DataFrame never needs).

Can public freight and trade feeds forecast warehouse demand?

The Census Monthly Wholesale Trade Survey adds the restocking signal: monthly sales, inventories and inventories/sales ratios across roughly 40 NAICS kinds of business within merchant wholesale trade, current through June 2026, delivered as multi-decade Excel workbooks of about 182 KB each under U.S. Government commercial delivery terms. The Census API needs a free key only for heavier programmatic use.

Want rows instead of a pitch? Name the datasets.

API, files, or your warehouse. Daily, weekly, or hourly.

Get a sample

Questions worth asking

How do I build a training set of industrial REIT fundamentals for free?

Use SEC EDGAR's XBRL companyfacts API: key-less JSON keyed by CIK holding every reported fact - 586 us-gaap concepts for Prologis with revenue facts from CY2016 through CY2026Q1 and per-fact filing references. Declare a User-Agent, stay near 10 requests per second, and cache locally; large filers run about 3.5 MB of JSON apiece.

Which series should I benchmark industrial REIT strategies against?

The FTSE Nareit US Real Estate Index Series on reit.com: 21 sector Excel files plus two index-level files refreshed monthly, annual returns beginning in 1972, and multi-decade monthly history including a dedicated Industrial file. Join Nareit's T-Tracker - about 274 variables across 107 quarters - for FFO, NOI and occupancy targets. Confirm FTSE Russell rights before republishing raw values.