What an index is
An index represents a group of assets as a single figure, so a broad segment can be followed through one value rather than many separate prices. The group may be a section of a stock market, a basket of commodities, or a set of currencies. What distinguishes an index from a simple basket is its methodology: a fixed, published set of rules defining which assets are included, the weight assigned to each, and the schedule on which the composition is reviewed.
What index data contains
The published value is only part of an index dataset. Complete index data documents everything required to understand and reproduce the value, which is what makes it suitable for benchmarking rather than casual reference.
- Value: the index level, as a time series.
- Constituents: the assets currently included.
- Weights: how much each constituent contributes.
- Methodology: the inclusion rules and the rebalancing schedule.
An index is a measure, not a tradable asset
An index cannot be bought directly, because it is a calculation rather than an instrument. The products that people trade are designed to track an index and behave differently from the index itself. For data purposes the distinction is important: an index value is a reference and a benchmark for comparison and context, not a price at which a transaction could have occurred.
Why methodology matters
Two indices covering the same market segment can move differently because they are constructed differently. Weighting each asset by its size produces one result, equal weighting produces another, and price weighting produces a third. Rebalancing rules determine how and when the composition changes. Because the result depends entirely on method, an index is a derived value, and the same underlying market can yield different index values under different rules. This is why the methodology is published alongside the value.
Building benchmarks on SiftingIO
SiftingIO does not publish branded indices, but it provides the underlying data required to construct and benchmark them. Normalized prices for thousands of symbols across stocks, forex, crypto, and commodities are available under one schema, so building a custom basket or comparing an asset against a segment is a matter of applying a weighting rather than integrating multiple data sources. The aggregated fair price is itself a rules-based composite, combining many inputs into one representative value under a documented method. Specific index datasets that are not yet covered can be requested.