Interactive Visualizations for Time Series Data
The ability to handle very large quantities of multivariate time series data is an essential element in a complete Visual Data Discovery solution. Datawatch offers a range of specialized data visualizations, including Horizon Graphs, Stack Graphs and Line Graphs, designed specifically to make analyzing historical data easier and more efficient.
Visualize data from any time series source
Datawatch's ability to connecto to columnar tick databases like Kx and OneTick as well as standard relational databases and OData sources is key to supporting fast, responsive multi-dimensional analysis of time series data. Our time series capabilities are especially important for users in capital markets looking at risk, performance, trading activity, and other very large time-stamped data sets.
Combine time series historical data with real-time streaming feeds
Datawatch dashboards can display interactive visualizations that combine time series data with other sources, including real-time message buses and the output of Complex Event Processing (CEP) engines. You can also program your own calculations using simple Excel-like formulas that use data from multiple sources to create new data columns. With this capacity, you can understand what is happening in the moment based on your real-time and CEP sources, and you can get a grasp on how you got to that point based on historical analysis.
In-memory data model supports responsive interactivity with time series data
Our StreamCube™ in-memory data model is designed for speed. This means our dashboards do more than simply display information. You can interact with the data on-the-fly. Slice and dice even extremely large data sets in a real-time environment. The system responds almost instantly to new filter selections and changes in the data hierarchy. For example, in a capital markets application, you can build a dashboard designed to analyze performance and risk for a portfolio of equities; StreamCube allows you to instantly change from a hierarchy based on geography to one based on industry sector. This helps you spot underlying patterns and outliers that might be hidden in a single dimensional view of the data.
Exploratory analysis of historical tick data
Datawatch can display time series data beautifully but it does far more. In applications like market risk analysis or algo-trading performance analysis, it's critical that the user be able to interact and dig down as needed to fully understand what has happened over the previous minutes, hours — or even weeks or months!
Tick data is voluminous. Datawatch can handle virtually any amount of data required to support historical visual analysis of large blocks of financial instruments for any amount of time. In order to support the interactivity required for exploratory visual analysis, Datawatch's ability to connect to tick databases is a key element in successful implementations.