Investment advice is a regulated category.
Investment advice is a regulated category. A personal recommendation is not just any market commentary. It is a communication presented as suitable for a person, or based on that person’s circumstances, in relation to a specific investment or a specific action concerning an investment. For publishers of quantitative research, that boundary matters because research can move from general analysis into a regulated recommendation if it is framed for an identifiable recipient or as appropriate for that recipient.
For Sonar Sciences contributors, the practical requirement is to treat publication as a compliance process as well as an editorial process. Sonar’s research-to-publishing guidance states that contributors are responsible for ensuring their content complies with applicable laws, regulations, and internal publication standards before submission and publication. It also states that content must be accurate, balanced, and appropriately caveated, and that contributors must not make misleading or unsubstantiated claims. Those requirements are directly relevant when a piece of research could be interpreted as investment advice or a personal recommendation because classification affects what can be said, how it must be framed, and what legal restrictions apply.
The core mechanism is classification. A publisher must first determine what kind of communication is being made. General educational material, methodological explanation, and broad market research are not the same as a personal recommendation about a specific investment. Once a communication is classified as investment advice or a personal recommendation under the applicable regime, legal boundaries attach to that classification. In practice, publishers should therefore disclose the regulatory character of the communication and the limits that apply to it, rather than leaving readers to infer whether the content is general research or advice tailored to a person.
Sonar’s publishing rules reinforce this approach through standards on evidence and presentation. The research-to-publishing guidance requires that factual assertions be supported, methods be described clearly enough for readers to understand the basis of the work, and limitations be stated where relevant. It also prohibits deceptive presentation of results. Those rules support a compliance posture in which a publisher distinguishes analytical research from advice and explains the basis, assumptions, and constraints of any claims made.
Performance reporting is one area where that boundary becomes concrete. Sonar’s backtest overfitting audit tool is designed to evaluate whether a reported strategy result may be overstated because the research process selected a specification that fit historical data too closely. The tool explains backtest overfitting as the risk that repeated testing and selection can produce an apparently strong result that does not reflect a robust underlying effect. It is used to audit a backtest by estimating the probability that the selected result is an artifact of the selection process rather than reliable evidence. In publishing terms, that helps substantiate compliance with standards against misleading claims because it adds a formal check on whether reported historical performance may be the product of over-optimization.
The deflated Sharpe ratio serves a related function. Sonar’s glossary explains it as an adjusted form of the Sharpe ratio that accounts for multiple testing and non-normal returns when evaluating whether an observed Sharpe ratio is statistically meaningful. This matters because a raw Sharpe ratio can look persuasive even when many variants were tried or when return distributions make standard significance assumptions unreliable. By using the deflated Sharpe ratio, a publisher can show that performance evaluation has been adjusted for selection effects and distributional issues. That supports more careful reporting and reduces the risk that published research presents backtest results with unwarranted confidence.
Taken together, these tools do not determine whether content is legally classified as investment advice, but they do help support the compliance obligations that arise in the publication of investment research. They provide evidence that performance claims were examined for overfitting risk and statistical inflation, which is consistent with Sonar’s requirement that research be accurate, balanced, and not misleading.
The key publishing lesson is straightforward. If a piece of content offers a personal recommendation about a specific investment, the publisher should identify the regulatory character of the communication and state the legal boundaries that apply to it. If the piece is instead general research or educational analysis, it should be written and presented in a way that does not imply individualized suitability. In both cases, Sonar’s own standards require substantiation, methodological clarity, and careful caveating of limitations.
Covered in depth in the From research to publishing pillar hub.