CETIN analysis methodology

Data, candles and engine

The backend obtains price series from external sources. The engine uses available data to calculate technical and contextual components. Resampling and timeframes do not add new market observations; insufficient coverage must remain visible.

Score, bias and confidence

The CETIN Score is a confluence indicator on a scale from 0 to 100. It summarises available factors, their agreement and their divergence. A score of 70 is not a 70% probability of profit. Bias expresses a calculated orientation at a point in time; it is not a universal instruction to buy or sell.

Confidence is an internal indicator of quality and coverage, not a historical success rate. Missing components must not be treated as favourable. No independent study establishing a success rate for the CETIN Score is published in the audited documents.

Timeframe and timestamp

M5, M15, H1 and H4 are analysed separately. The current price time and the analysis time have distinct roles. Backend timestamps are in UTC; local displays should indicate the timezone. Check the source and age of the observation before interpreting a signal.

Macro and intermarket context

An economic event can change context or volatility. Its result does not determine the next direction with certainty. A correlation between instruments is not a permanent causal relationship. Opposing factors and missing data must remain identifiable.

Data status

LIVE: freshness checks passed, without a guarantee of a direct exchange feed. DELAYED: observed or declared delay. STALE: insufficient freshness. CLOSED: market identified as closed. UNAVAILABLE: no usable data. DEMO, when used, indicates a demonstration rather than a verified price.

A backup copy shown after a disconnection must display its actual age. A LIVE status does not remove the risks of provider delays, errors or differences from a broker’s prices.

Traceability and publications

To make an analysis verifiable, retain the exact instrument and product, timeframe, timestamp, price source, engine version, factors and limitations. Actual weights and thresholds, the version associated with each analysis and the change history must still be checked against backend sources before being published as facts.

Depending on its content and the instrument concerned, a public analysis may fall under investment recommendation rules. Lack of personalisation does not exclude that regime. Applicable sources, methods, date, author and conflicts of interest must be accessible alongside the publication.