Methodology
Last updated: November 12, 2026
GoldSignal Data organizes filings and reported ownership activity into source-linked research screens. Automated detectors and AI-assisted context help identify records worth investigating. The system provides evidence and context, not investment picks. Strategic ownership is the primary research focus; insider activity provides supporting context.
Data source tiers
- Filing-Sourced (SEC) Holdings from SEC 13F quarterly filings, Form 4 insider transactions, and Schedules 13D/G. Machine-parsed from EDGAR with verified CIK mappings.
- SEDI Canadian insider filings ingested for CSE, TSX, TSXV, and other SEDI issuers we cover. Named separately from SEC — a SEDI name is not a Form 4 / 13F filer.
- Third-party insider data (INK) Some insider trading and related ownership reference data is obtained from INK Research Corp. (INK), an independent financial information provider. It supplements public SEC and SEDI disclosures; source coverage and reporting dates may differ.
- Fund Disclosure Holdings from official fund fact sheets and regulatory disclosures published by the fund or its administrator (for example, HL/Broadridge). Not an SEC filing, but an official document from the fund itself.
- Curated Holdings from newsletters, interviews, podcasts, press releases, and public statements. Manually verified against the original source.
- Market data Prices, market caps, betas, and historical series from Polygon.io (primary). Company profile and fundamental fields may be filled from Financial Modeling Prep when the primary feed is thin, especially OTC and Canadian-listed junior miners. FMP is supplemental, not a co-primary real-time feed. Both providers are cached in the background; pages do not call them live on load.
Portfolio curation
- Investors are selected for precious-metals relevance; holdings are normalized to tickers and deduplicated across filing types.
- Positions and insider rows are validated for completeness (dates, sides, share counts) before they appear in dashboards.
- Updates follow new public filings; cached aggregates refresh on a scheduled cadence.
Automated and AI-assisted signal review
The research workflow starts with issuer identity, filing type, transaction date, investor role and source URL. Screens surface repeat buying, large positions and independent ownership patterns. AI-assisted narrative context can help summarize disclosures, but must not replace filing facts or introduce unsupported claims.
Missing capitalization, stale data, identifier ambiguity, or incomplete disclosures reduce confidence. Unverified results must be labeled or withheld, not filled with AI-generated estimates. A screen is an invitation to examine evidence, not a buy rating, return forecast or model portfolio selection.
Gold Beta (OLS regression)
Gold Beta estimates how much a stock's daily returns co-move with its benchmark, GLD for gold-focused names or SLV for names whose primary revenue is silver, using ordinary least squares (OLS) regression on aligned daily return series. Benchmark assignment is an explicit, human-maintained list keyed on each company's primary metal, not a keyword match on free-text sector data, and it is stored on the ticker rather than re-inferred on every page load.
For each stock, we regress stock return on benchmark return over the trailing 252 trading sessions of split- and dividend-adjusted closes. The slope is Gold Beta; the coefficient of determination (R-squared) indicates how much of the stock's day-to-day variation is explained by benchmark moves over that window.
Before regressing, we inner-join the stock and benchmark series on trading date (a date missing from either side is dropped, never zero-filled or forward-filled), drop days with zero or missing stock volume, drop days where the stock's return was exactly 0.0 while the benchmark moved (a stale price rather than a real no-move day), and drop single-day returns beyond 400 percent as unadjusted-split or bad-tick data. Large but plausible moves are kept and only logged.
We surface beta together with R-squared so you can judge reliability. Below an R-squared of 0.10, or with too little clean history or too many dropped days, we do not publish a beta at all: the stock shows as not computable with a reason. Between 0.10 and 0.20 the beta publishes but is flagged low confidence. We never apply a bias, floor, multiplier, or winsorization to make the published distribution look a particular way.
Reading the Radar
The public Radar is a reviewed coverage set, not a complete shareholder register. It shows dated equity stakes with holder types, ownership basis and original sources. Missing coverage does not mean no strategic investor exists.
Mining companies, royalty companies, investment funds and other strategic investors are labeled separately. A holder’s business type does not identify its investment instrument: equity in a company is different from a project earn-in, exploration alliance, royalty or stream.
Project rights and partnerships must be read in the underlying disclosure; they are not additional shareholdings. An equity stake does not establish the investor’s motives, independent technical diligence or an attractive valuation for public shareholders.
“Multiple backers” means more than one reported holder group in the same company. It does not require two 10% stakes, prove independence, or represent a tested investment signal. Stakes are shown separately rather than combined into a headline percentage.
“Recent changes” uses the ownership effective date within the past 30 days and a reported initial investment, increase, reduction or dilution. Adding an old filing to coverage does not make it a new purchase. Disclosures older than 180 days remain in history; that age label is not evidence of an exit.
Historical Strategic Convergence (SIG-030)
A historical screen of reported 10%+ holder overlap within 12 months. Independence and continuing ownership require verification.
The research hypothesis concerns two independent 10%+ holders on the same gold or silver company within 12 months. The archive runs through August 2026. It is distinct from the Radar’s descriptive “Multiple backers” view and is not a live predictive alert.
Historical in-sample findings are not evidence of future returns. Read the research record for scope, limitations and validation status; do not infer performance from a current holder count.
Read the research record · Strategic Convergence methodology & access
Concentration lists
- Built from reverse-looked-up 13F holders of the tracked universe. Banks, brokers, index managers, and generic wealth platforms are on an explicit exclusion list.
- Concentrated Bets: company market cap under $1B, and the position is among the institution's top 3 GoldSignal-universe holdings. One company per card (largest qualifying position). Positions under $500K are omitted.
- Specialist Overlap: held by two or more filers on the explicit precious-metals specialist list (for example Sprott, Euro Pacific, Adrian Day / Global Strategic, McAlvany). Banks never count.
- Small and Concentrated: under $1B, where the position is at least 2% of company market cap or is top-3 in that institution's GoldSignal-universe book. Dollar size, book rank, and mcap percentage stay separate.
Data sources & coverage
GoldSignal combines public records, third-party market data, and derived research indicators. Source coverage differs by company, investor, jurisdiction, and date. The presence or absence of a record is not a complete account of an investor's holdings or an issuer's activity.
- Regulatory ownership and insider records: U.S. SEC EDGAR (including Forms 13F and 4 and Schedules 13D/13G) and Canadian SEDI insider disclosures, where covered.
- Third-party insider data: Selected insider-trading records and related reference fields from INK Research Corp. (INK) may be used alongside public filings. INK is an independent provider; its data may be delayed, incomplete, corrected, or subject to different coverage and definitions. Check original SEC or SEDI filings before relying on a transaction.
- Company and investor disclosures: Issuer news releases, regulatory announcements, corporate presentations, official fund reports and fact sheets, and source-attributed investor statements, interviews, or publications. Curated public statements are not equivalent to regulatory filings.
- Market and company reference data: Prices, trading volume, price histories, market capitalization, and selected company and fundamental fields from Polygon.io (primary) and Financial Modeling Prep (supplemental), where available. Feeds may be cached or delayed.
- GoldSignal calculations and classifications: Ownership changes, concentration, convergence, relative momentum, and beta are calculated or classified from available underlying records. They are research interpretations, not independent primary-source facts or guaranteed signals.
Important disclosures & limitation of liability
Information only; no investment advice. GoldSignal Data provides general research and education, not personalized investment, legal, tax, or financial advice, a recommendation, or an offer to buy or sell securities. No user should treat a displayed position, insider transaction, signal, ranking, or commentary as an endorsement of a security or a prediction of future performance.
Accuracy and completeness are not guaranteed. Information may be inaccurate, incomplete, inconsistent, duplicated, misclassified, unavailable, delayed, or outdated. Original filings and third-party feeds may themselves contain errors, omissions, later amendments, or conflicting figures. Automated extraction, entity and ticker matching, aggregation, calculations, editorial review, and AI-assisted summaries can introduce additional mistakes. We do not guarantee that all issuers, investors, transactions, filings, or changes are captured, or that identified errors will be corrected within any particular period.
INK Research and third-party provider disclaimer. GoldSignal sources some insider-trading and related information from INK Research Corp. (INK). Naming INK as a source does not mean that INK has reviewed, approved, sponsored, or endorsed GoldSignal, its calculations, rankings, commentary, or conclusions. INK is not providing investment advice through GoldSignal. INK states in its own published disclaimer that it does not guarantee information accuracy or completeness and is not responsible for users' investment decisions. Source-provider terms and restrictions continue to apply. To the maximum extent permitted by applicable law, GoldSignal and its independent data suppliers, including INK, disclaim warranties and liability for inaccurate, incomplete, delayed, or misinterpreted information and for losses arising from reliance on it. Nothing in this notice excludes liability that cannot legally be excluded. See INK's published disclaimer.
Disclosed ownership is not live ownership. A Form 13F generally reports certain holdings as of a past quarter-end and is filed after that date; it does not show every asset, short position, trade, or present-day holding. Schedules 13D/G, insider filings, fund reports, and public statements have different scope and reporting dates. A reported stake may have been sold, increased, diluted, transferred, or otherwise changed without that change appearing immediately on GoldSignal. Absence of a reported holding does not prove absence of ownership.
Market data and analytics are estimates or snapshots. Quotes, volumes, market caps, valuation inputs, corporate-action adjustments, currency conversions, and historical series may differ across providers and trading venues. Calculated indicators depend on source availability, assumptions, and chosen time windows; they can change when underlying data is updated. GoldSignal is not an execution-quality, real-time market data service and should not be used to place trades.
Independent verification and risk. Users must independently review original disclosures, current market conditions, and relevant financial information before acting. Mining and exploration securities can be volatile and illiquid and may involve financing dilution, operational setbacks, permitting uncertainty, commodity-price exposure, political risk, and loss of all invested capital. Users are solely responsible for their investment decisions and should consult appropriately licensed professionals when needed.
No warranties; limitation of liability. All content and services are provided "as is" and "as available," without representations or warranties of accuracy, completeness, timeliness, reliability, fitness for a particular purpose, or uninterrupted availability, to the fullest extent permitted by applicable law. To the maximum extent permitted by law, GoldSignal Data and its owners, operators, contributors, and data suppliers disclaim liability for losses, damages, costs, or claims arising out of or relating to access to, use of, inability to use, or reliance on the site or its data, including trading or investment losses and losses arising from inaccurate, incomplete, or delayed information. Nothing here excludes liability that cannot legally be excluded.
For additional conditions governing use of this website, see our Disclaimer and Terms of Use.
Educational information from public filings. Not investment advice.