Introduction

The Exabel Signal DSL enables the user to work with time series data in a new way, which is both intuitive and powerful.

Our first tool using signals is the Signal Explorer, which allows the user to visualize time series, including transformations of the raw data series, and build statistical models of these time series.

A signal is a time series, possibly parametrised by company, which conveys some information. A signal can either be raw data, or it can be a calculation involving other signals.

The GDP of the United States and the price of Brent crude oil, are examples of raw data signals which are not parametrised by company.

An example of a parametrised raw data signal is revenue. The revenue for Ford Motor Co. is one time series, while the revenue for General Motors Co. is another. Another example of a parametrised signal is the metric P/E, which is calculated by dividing the market cap of the company by the earnings of the company.

A calculation is specified by a formula, which can combine multiple signals to yield new signals. Exabel has developed a powerful way of expressing such formulas in a generic way, which lets you apply the same formula to any company. This allows you to express advanced calculations or analyses in a concise way, and apply them to tens or hundreds or even a thousand companies, and compare the results across all of them. This allows you to benchmark companies quantitatively.

Signals can be forecasted into the future. For example, the expression transactions.forecast('theta') yields a signal which is identical to transactions as long as that signal has data, but extends it into the future by calculating predictions based on the historical values. Such predictions can then be visualized, monitored, and even used as input to other models.

You can use some signals to predict another signal, with the help of Exabel’s auto-modelling technology. The prediction becomes a new signal.

Thanks to the powerful combination of cloud computing, expressive formulas and auto-modelling technology, quantitative analysis can be applied at scale to monitor the market for insights and investment opportunities.