S1 Primary
Filings, official statistics, regulatory records, direct company disclosures, court records, and government data.
Research standards
Sources, uncertainty, updates, and corrections.
A useful repricing thesis must show what is known, what is inferred, what would verify it, and what would prove it wrong.
A thesis that cannot be disproved is marketing.
Source hierarchy
S2 Specialized data
Trade, pricing, inventory, shipping, capacity, backlog, policy, and industry datasets with a defined methodology.
S3 Reliable reporting
Reporting from reputable financial, technical, or industry sources used as leads or context, not as final proof of a thesis.
S4 Weak lead
Headline-only, secondary, incomplete, or low-context material. These items can start a watchlist but cannot carry a published thesis.
Claim hierarchy
E1 Fact
Disclosed or directly observable facts, such as a filing, order, project milestone, official statistic, or published rule.
E2 Relationship
Established industry or market relationships, such as demand units, conversion ratios, capacity constraints, or process dependencies.
E3 Inference
Open Variable Research's interpretation of how a delivery chain, buyer, operator, supplier, or regulator may respond.
E4 Scenario
Conditional repricing logic. These claims depend on verification signals appearing and must be withdrawn if disproof signals appear.
What a complete signal must include
- A dated fact pattern or current market signal.
- A concrete core demand unit.
- The delivery chain required to satisfy that demand.
- The scarce, slow-to-expand, policy-sensitive, or margin-changing mechanism node.
- The proposed capital relevance, separated from any valuation or underpricing conclusion.
- At least three verification signals.
- At least two disproof or withdrawal signals.
- A clear evidence standard and next verification path.
Updates
Material changes to the evidence, verification signals, disproof conditions, or status of a public thesis should be reflected in the relevant research page or signal record. A revision clarifies what changed, why it changed, and whether the research status moved.
An old topic may be reused only when a dated new fact changes the mechanism, evidence, verification path, or disproof path. Open Variable Research should not present a repeated topic as if readers are seeing it for the first time.
Corrections
Open Variable Research corrects material factual errors and distinguishes a factual correction from a change in interpretation. Send correction requests, the source, and the affected page to julian@nexvale.io.
Correction requests should identify the page, claim, proposed source, and why the correction changes the fact record. Disagreement with an interpretation may lead to an update or response, but it is not the same as a factual correction.
AI use boundary
AI tools may support organization, drafting, structured extraction, and visual production. AI output is not treated as primary evidence. Public claims must still be grounded in sources, labeled by claim layer, and reviewed before publication.