Product description: Our "Lighthouse" library contains prognostica’s know-how collected over the years about time series forecasts in one place. The software programmed into Java has managed to transfer our experience gained from R prototypes into a productive and professional structure. Compliance with modern software development standards ensures its high quality.

The forecasts are based on time series data and can be implemented for any forecast horizon. Among other things, our algorithms can take into account up to three seasonal factors (e.g. annual, monthly, and daily seasonality) as well as a range of economic indicators or other influencing factors (see Indicator Matching). The library can deal with levels of time granularity from years to weeks to hours. In addition to the actual prediction, our methods also provide information on the model’s quality as well as forecast intervals with best-case and worst-case estimates, thus allowing for a more realistic assessment of the results.

This is what we do for you: Working together with your team, we define the integration of our "Lighthouse" library into an existing system or into a new joint product. Afterwards, we design an application programming interface (API) with the desired functionality to meet your exact requirements. That’s how you can benefit from our comprehensive expertise in the field of time series forecasts.

What you contribute: A well-maintained database. If necessary, your own forecasting program, which you want to extend.

Target group: What we offer is especially aimed at companies that want to design a joint product with us or extend an existing product with the help of "Lighthouse".

Product specification: If the influencing factors associated with a time series are known and are also predictive or leading in nature (so-called leading indicators), the development of a business can be extrapolated. This allows our customers to act proactively: They are prepared for a downturn in sales because of a worsening economic climate. They can react appropriately to an imminent increase in sales with an expansion of resources and respond profitably to an economic upswing.

Consequently, our "indicator matching" service is attractive to companies whose business is closely intertwined with the current economic situation. This is often the case with large companies or corporations with a large range of products or with companies whose products make up a large portion of the respective market. Additionally, companies whose revenues are significantly driven by end-user business and are therefore closely connected to consumer behavior can profit from our service.

Even if the leading character of the influencing factors cannot be verified, one thing is none the less gained: the knowledge of which other areas of business or influencing factors exhibit a direct or indirect connection to one’s own business.

This is what we do for you: We examine the relationship between your time series and a multitude of economic indicators or influencing factors we have defined ourselves. After this identification of the factors that are connected to your time series, those that have predictive properties (so-called leading indicators) are identified. We compile a list with those influencing factors that are relevant to you, as well as the kind of correlation identified, such as its impact, direction, and time leading.

In a consultation, we jointly define a reasonable analysis level as well as criteria for pre-selection of potentially relevant influencing factors. You are welcome to introduce possible influencing factors yourself, for instance prices of raw materials, competing products, or sales figures of adjoining business segments.

What you contribute: one or more monthly time series—ideally without values missing—with at least a five-year history (e.g. monthly revenue of the whole company or the business sector); if necessary, your own possible influencing factors; other forms of granularity (e.g. data on a daily, weekly, or yearly basis) are available on request.

Target groups: B2C companies, B2B companies

What it costs: The price of our service is calculated by the amount of data to be analyzed as well as the number of possible influencing factors investigated. Feel free to contact us!