Data.gov Securities Portal

Data.gov Securities Portal is the Investment Banking & Brokerage catalog's discovery record over the United States government's master index of its own data. Run as a General Services Administration property, the federated catalog advertised 552,271 dataset records at verification, harvested from federal agencies plus state, county and city publishers. Dozens concern securities directly: SEC and Treasury collections covering fund portfolios, broker-dealer supervision and market statistics.

The headline entry is the SEC Division of Economic and Risk Analysis harvest of Form N-PORT Data Sets - 27 quarterly ZIP packages spanning 2019 Q4 through 2026 Q2, each carrying portfolio holdings, asset categories, derivatives, securities lending and fund-level financials for registered investment companies, with the publishing program declaring a quarterly release rhythm in ISO 8601 terms. Around it sit Treasury interest-rate and debt records and the Census Bureau's 293,639-record facet, which together indicate how deep the federal layer runs.

One caveat shapes everything else: the catalog indexes descriptions of datasets, not the datasets themselves. Every record points at its publisher. Get a sample of this dataset and Datadory hands you the resolved securities slice - fields populated, rows delivered - instead of a reading list.

Two verified records ship with the page, exactly as the catalog describes them:

title        : Form N-PORT Data Sets
organization : Economic and Risk Analysis | Securities and Exchange Commission
resources    : 27 quarterly packages
media_type   : application/zip
cadence_code : R/P3M

title        : Electric Vehicle Population Data
organization : State of Washington
formats      : json xml csv kml html

Read the pair together and both halves of the proposition appear. The first row is the reason securities desks care: a single record enumerating 27 quarterly fund-disclosure packages, typed as ZIP distributions, cadence-coded for quarterly arrival. The second row is the honesty layer - ask the catalog for securities and it answers with everything government publishes, electric-vehicle registrations included, because discovery here is free-text over the whole corpus rather than a curated shelf. Knowing which records deserve your attention is the actual product, and it is why the resolved slice beats the raw pointer list.

Six fields form the verified core of every record. Two identify it - title as the publisher wrote it and organization as the accountable bureau - and dataset_last_updated stamps when the record last moved, down to the minute. The three distribution fields carry the engineering weight: downloadURL locates each described file on the publisher's host, mediaType types it, and accrualPeriodicity preserves the publishing program's declared rhythm as an ISO 8601 code.

Eight further field families were observed across the slice during verification and fold under additional fields on request: result-card abstracts, format badges, catalog-check timestamps, keyword and organization facets, geographic box filters, the downloadable-file flag and the sort keys. All are parts of the same envelope rather than guesses - name the ones your workflow touches when you request a sample and they arrive as populated columns.

Geography - the United States throughout, at unusual spread. Federal agencies supply the securities-relevant core - SEC and Treasury above all - while state, county and city publishers fill the surrounding index. A regulator's record and a state registry parse identically.

Temporal - governed entirely by the underlying publications. The flagship N-PORT record spans quarterly packages from 2019 Q4 through 2026 Q2; elsewhere each record carries its own movement stamp, the verified example reading October 08, 2024 at 03:22 PM. There is no catalog-wide freshness promise to lean on - the per-record timestamp is the truth.

Granularity - one record per published dataset, stepping down to resource-level distributions inside each record. The N-PORT entry alone itemizes 27 quarterly resources, which makes it a small archive in its own right.

Set against the wider Datadory catalog - where the average quality score across all 1,744 datasets is 7.81 - this record scores 6/10, capped by index depth rather than documentation: the six-field dictionary is fully verified, but a discovery layer cannot match a primary statistical series for analytical density.

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

Channel and cadence are settings, not projects. A nightly warehouse load keeps fund-holdings screens current between the quarterly N-PORT arrivals; weekly files suit analysts who benchmark once a cycle; structured feeds suit products that surface a securities record inside an app. Every delivery travels with the field dictionary above, the sample rows and the coverage profile mapped to the agencies and collections you named - so the schema you see in the sample is the schema you ship against.

• Investors & quant researchers - work from primary SEC filing packages discovered through the catalog rather than third-party aggregators, with N-PORT quarters landing as filed (investors & quants use cases).
• Data scientists & ML engineers - feed fund-holdings pipelines from complete quarterly archives whose metadata keys arrive attached, skipping the hand-collection stage (data scientists use cases).
• Developers & data-product builders - build discovery tools on a uniform record envelope: one parser covers federal, state and city publishers alike (developers builders use cases).
• Journalists, academics & students - anchor stories and papers to official federal records with per-record timestamps, citable without vendor mediation (journalists academics use cases).
• Market researchers - size US financial-market topics from federal tables before commissioning anything niche (market researchers use cases).
• Competitive-intel product teams - check whether regulators already publish on a segment before paying for custom collection (competitive intel product teams use cases).

Investors and quant researchers rank highest (relevance 3 of 3 in Datadory's tagging): primary-package sourcing beats aggregator drift when a thesis rests on what funds actually reported. Journalists, academics and students (3/3) get citation-grade federal records whose per-record timestamps survive editorial scrutiny. Data scientists and ML engineers (2/3) inherit keyed, typed records instead of maintaining hand-built collection code. Developers and data-product builders (2/3) wrap products around a stable six-field envelope. Market researchers and consultants (1/3) and competitive-intel product teams (1/3) use the catalog as a due-diligence step - confirming what the public record already holds before anyone spends budget collecting it.

Within investment banking and brokerage data - twelve primary datasets in the Datadory catalog - this record owns discovery: it is the router over the entire federal layer. The neighbors own jobs. U.S. Treasury Interest Rate Data (Daily Yield Curve) prices time across fourteen tenors back to 1990 and scores 10/10. Federal Reserve Bank of New York Markets Datasets documents central-bank operations and dealer positions. FINRA API Developer Center reaches TRACE and short interest; MSRB EMMA covers municipal trades; SIFMA Capital Markets Fact Book reports annual volumes; WhaleWisdom 13F Aggregator rebuilds institutional ownership as one table; Jay Ritter IPO Data carries issuance history back to 1975.

When you know the series you need, go direct - the specialists are faster and deeper. When you need to learn what the United States government publishes on securities at all, this is the front door.

Three things worth having in hand.

First, this is a map, not the territory. Records describe datasets their publishers hold; nobody downloads a table from the catalog itself. Datadory's job in this record is resolution - turning pointers into delivered rows with the dictionary intact.

Second, search behavior has a quirk. At verification, a bare securities query returned the catalog's default most-popular listing rather than a filtered set, and no curated securities topic page surfaced. Explicit terms, organization facets and the downloadable-file flag are what isolate the SEC and Treasury material - which is precisely the trimming a resolved sample performs for you.

Third, verification went record-by-record, because the catalog's legacy bulk-query interface answered 404 during the August 2026 pass. Coverage stated here was confirmed on visible records rather than asserted from aggregate counts, so treat the slice boundaries as empirically drawn rather than promised.