Indicators

DataCat can calculate Registry-reviewed technical indicators while it writes an OHLCV dataset. The catalog on this site is generated from the same governed Registry used by the app, while each indicator article adds interpretation and usage guidance.

Types and instances

An indicator type is the reviewed calculation definition. An Indicator Instance combines that type with resolved parameter values inside an Indicator Set or frozen Job configuration.

You can use multiple instances of the same type. This is useful for comparing horizons, but each instance adds outputs, warmup requirements, calculation work, and dataset columns.

Categories

Categories help browse calculations with related purposes, but they are not a guarantee that two indicators are interchangeable.

1. Overlap Studies generally produce smoothed or modelled values in the scale of price.
2. Momentum Indicators describe direction, rate, relative position, or trend strength.
3. Volume Indicators combine or accumulate price and volume information.
4. Volatility Indicators describe range or dispersion rather than direction.
5. Statistic Functions apply rolling statistical models to the input series.

The Website generates category names, counts, and indicator links from the validated Registry-to-article join rather than maintaining a second catalog here.

Choosing an indicator

Start from the analytical question:

1. Do you need a smooth baseline, rate of change, range, dispersion, trend strength, volume confirmation, or a regression feature?
2. Should the output preserve price units, use a percentage scale, or be unitless?
3. How much warmup can the workflow tolerate?
4. Will repeated parameter variants add distinct information or only correlated columns?
5. How will pending, omitted, and unavailable source bars be handled downstream?

Do not treat a familiar overbought/oversold threshold or crossover as a universal prediction rule. Indicator meaning depends on timeframe, market regime, provider data, parameterization, and surrounding features.

Parameters and structured facts

Each indicator page receives its title, category, description, inputs, parameters, accepted bounds/options, outputs, and implementation facts from the Indicator Registry during the Website build.

The long-form article does not repeat those facts. This keeps the app catalog and public documentation aligned when reviewed metadata changes.

Warmup

Many calculations require earlier bars before the first output. DataCat persists NULL for warmup rows and marks the output state as warmup.

Warmup is derived for the configured Indicator Instance; it is not a single static number that can be documented correctly for every parameter choice.

Data-quality states

An output can be final, provisional, derived, warming up, awaiting input, or unavailable. If your analysis requires only durable provider-backed values, filter both the source provider_data_status and the effective indicator output state.

Comparing results with another tool

Differences can come from input provider, timeframe boundaries, parameter defaults, moving-average type, warmup convention, missing-period handling, precision, or calculation engine version.

Before comparing values:

1. Match the exact symbol, provider, timeframe, and UTC boundaries.
2. Match every resolved parameter.
3. Compare only rows where both systems have sufficient warmup.
4. Check whether missing periods were omitted, filled, or normalized.
5. Use the dataset’s indicator_definitions and version metadata instead of relying on a UI label alone.