Radar Perene / Archive / science

Adapt or import: the decision that precedes any Brazilian sector index

◦ Index methodology v2.2 (working papers with DOI). See the methodology.

Science

There is a moment, in every piece of research applied to a small market, that appears in no final paper: the moment someone decides whether the foreign formula comes in as is or needs to be rebuilt. It is a silent fork. Whoever imports without asking saves months and inherits broken premises; whoever adapts without criteria trades the original's defects for the defects of improvisation. The quality of everything that follows — the series, the index, the conclusions — is decided in that instant, before the first line of code.

Before applying any imported methodology, the house submits the formula to a formal decision: import as is, adapt, or do not use — and documents the decision, with its justification, alongside the study's material. This is not bureaucracy; it is the recognition that the choice between importing and adapting is itself a research result, and undocumented research results can be neither audited nor revised.

The previous articles in this trail showed why: the source literature assumes conditions Brazil does not have, and the house's work with sector baskets met those breaks in practice, not in theory. This article is about what one does with the diagnosis.

Why the decision must be explicit

The alternative to the explicit decision is not neutrality — it is the implicit decision, taken by inertia and invisible forever. When a team applies the American formula "because it is the standard", it has decided to import; it just did not record that it decided, or why, or against which alternatives. Years later, when the series behaves strangely, no one knows whether the problem lies in the market, in the formula or in the premise no one audited — because the audit never existed.

The written record changes the nature of error. A documented adaptation that proves insufficient is correctable: one knows what was changed, for what reason and which premise the change addressed. A silent importation that fails is a mystery: the defect can live in any layer. The difference between the two scenarios is not the error rate — it is the cost of finding the error.

The three outcomes, and what each demands

The cheapest outcome — import as is — is legitimate when the formula's premises survive the local environment, and that survival was verified, not presumed. Some techniques depend little on market structure; adapting them on principle would be theater.

The second outcome — adapt — is the most frequent in the house's sector work, and the most delicate, because adaptation without discipline opens the way to the opposite vice: shaping the formula until it produces the result one expected from it. The protection is the same that applies to any methodological choice: the adaptation must be justified by the broken premise, not by the desired result, and it must survive the screen Track 1 describes in robustness testing — if the adapted version only works in one exact configuration, the adaptation created an artifact, not an instrument.

The third outcome — do not use — is the least published and the most informative. Some methodologies simply have no object in the Brazilian market, and the honest decision is to record that and proceed without them. The house's internal records hold decisions of this kind, and they cost little: the methodology discarded at the entrance generates no bad series, no fragile conclusion, no future retraction.

The governance layer

Documenting the adaptation decision is not an isolated habit — it is one piece of a larger principle the house applies to its own production: relevant methodological choices leave an auditable trace. The full criterion, point by point, is internal operation; the principle is public and verifiable in its effects. When a house study is deposited, the version records what changed; when a methodology fails, the correction becomes a dated rule; when an imported formula goes into production, there is a record of why it entered in that form. None of this improves a formula by itself — it improves the ability to discover, later, where the formula was worse than it looked.

It is also the link between this trail and the next. Deciding how a methodology comes in is half of governance; the other half is deciding what of the result becomes public — under which identifier, with what degree of openness, with which declared limits. That is the question the house's published indices answer, each in its own way, starting with the next article.

Frequently asked questions

Doesn't adapting a methodology make it incomparable with the literature?

It makes the comparison more laborious and more honest: results are compared knowing what changed and why. The alternative — perfect comparability over broken premises — produces aligned numbers that measure different things.

How can one tell an adaptation has gone too far?

The classic sign is the justification migrating from the premise to the result: when the change is explained by "it works better this way" rather than "this way the local premise is met", adaptation has become fine-tuning of the outcome. Retesting under variations is the most reliable detector.

Should all applied research document this decision?

The cost is one page; the benefit appears at exactly the worst moment — when something fails and one needs to know where to look. Teams that do not document have not decided less; they have decided invisibly.

Why does the house not publish the full decision criterion?

Because the operational criterion is part of the craft that sustains the house's production. What matters to the reader is public: the existence of the formal decision, the three possible outcomes and the verifiable effects in the deposited material.

---

Continue the trail: Perene Risk: the index with a DOI and no public formula — and why that is the right choice →

House reading: today's reading is in the Diário; the patterns that survived adaptation, in the Atlas.

Assessing whether an imported methodology fits a specific case — and what would need to change for it to fit — is an exercise the house conducts in conversation.

Characters: Method