For your convenience, Quandl has optimized our datasets for one of two different formats: 1) Time-series and 2) Tables.

📘

NOTE:

To find out whether a dataset supports time-series or tables, simply go to the Data section of its documentation and look under API.

For example, the FRED dataset shows an API type of Time-series, whereas the Mergent Global Fundamentals dataset shows an API type of Tables.

As a data platform, we take great pains to consider the best shape to format our different datasets. Currently, most of our datasets are accessible in either time-series format or table format. A few of our datasets, however, are accessible in both time-series and table format. Consequently, it is important to understand the differences and overlaps between the two data formats.

The Time-series API

A time-series is a collection of observations or measurements taken over a period of time, generally in equal intervals. Time-series are commonly displayed using line graphs, where the X-axis represents dates and the Y-axis represents other numeric observations.

It is important to note that time-series only contain numeric data types and are indexed by one date field. In other words, time-series data are always sorted by date. Through our API calls, users can retrieve the entire time-series as well as a slice of it.

Most of Quandl's datasets are stored as time-series because financial data generally consists of two types: dates and observations, which perfectly fit the time-series format.

The Tables API

While time-series only contain sorted numeric values, tables can include various unsorted data types (strings, numbers, dates, etc.) and can be filtered by different fields.