FAQs

I. What do I get if I enter into a contract with prognostica?

You will receive tailor-made solutions to suit your individual needs. Depending on the scope of the project and the project objective, we can approach your question in different ways: Our core expertise is our profound methodological knowledge, which we exploit within our consulting services either in the form of workshops or short-/mid- or longterm projects. As part of our consulting service, we are happy to create individual program code, graphics, reports, etc. for you and your needs at your request. We usually provide you with the appropriate program code within an operationalization project.

II. What does prognostica’s licensing model look like?

The program code we provide (e.g. as part of an operationalization project) can usually be used and developed permanently within your company. We do not allow it to be distributed to third parties. If you desire maintenance or further development of the code or if you would like to receive program code/software updates from us, further consulting costs will be incurred, which we will be happy to discuss with you.

III. Why do you recommend an initial workshop before the "real" project?

Particularly in the case of innovative predictive analytics projects in the evaluation and conception phases, the exact procedure and the associated effort involved in approaching a previously set vision is a priori difficult to grasp and define in detail. However, we are firmly convinced that a certain degree of agility is needed right at the beginning of the project in order to fully unleash its potential. The fastest possible adaptation or even rejection of the chosen path and its alignment are essential at the beginning. Through an intensive initial workshop, we can gain an early sense of the potential in your data and target a suitable project direction.

IV. Do we need to provide you with data?

It is very important that we get a realistic impression of your data. Often this is only possible if we can also analyze real data. If it is not possible for you to provide us with such data, it may to some extent be possible for us to work with sample data that is structurally close to the real data. However, this must be verified in each situation.

V. Which data formats can you handle?

The usual data formats pose no problem to us: csv, txt, xlsx, RData, data from relational databases (SQL), JSON ... But the list is not exhaustive: Do not hesitate to ask about the compatibility of your data format.

VI. Where and in what form does the calculation of the results take place?

We are flexible and will gladly implement your request: The calculation of the results can take place either on your digital infrastructure or on our computing cluster. Cloud services are also conceivable.

VII. What happens to the data we provide for calculations?

The data will be stored and processed on encrypted computers solely for the purpose of fulfilling the terms of the contract. Usually, the data will not contain any personal information. Your data will be treated confidentially, used only for its intended purpose, and deleted as soon as possible.

VIII. What software do you use? In which programming language is your code written?

Our program code is usually written in R, Java, or Python. On request, we can also help you in other programming languages.

IX. In which format do we get the results?

We are flexible in processing and tailoring the output according to your wishes: The results can be stored in a database, saved in a csv format, issued as a PDF report, or made available to you via a user interface.

X. Can we actually "see" the results?

Of course, numbers and code are often hard to comprehend. Therefore, it is always a good idea to visualize the data. We are happy to create a user interface, dashboard, or report that will allow you to view the results in a user-friendly way.

XI. Can you also use machine learning or do you use artificial intelligence?

Yes. Our repertoire of prediction methods includes various machine learning methods, such as multivariate adaptive regression splines, support-vector machines, classification and regression trees, or neural networks. We have extensively tested their application in a time series context.

XII. Is it possible to buy (economic) indicators from prognostica?

Currently, we do not sell any indicators. Many economic / industrial indicators are freely available. Others can be purchased from third-party vendors if needed. We will gladly evaluate with and for you which indicators are suitable for your project.

XIII. What you do ... can’t Excel do just about the same thing?

At first glance: yes. At a second glance: no. Time series forecasts can also be calculated in Excel. However, we offer many additional forms of functionality, such as the inclusion of additional information (weather, public holidays in daily forecasts, economic indicators), the optimization of a forecast in terms of a larger forecast horizon, machine learning algorithms, but above all the intelligent selection of the appropriate methodology for each application. In addition, we do not neglect what happens before and after the forecast: We handle both the preparation of the data—so that they are suitable for forecasting—as well as appropriately report the forecasting results for you, to name a few of the differences.

XIV. Does the information we provide need to meet certain requirements?

Data quality is critical and fundamental to the quality of all our analyses. There is no minimum amount of data. Depending on the application, we can handle both small and large amounts of data. In many problems presented to us we deal with time series data, which meet the special requirement that they are recorded at regular time intervals and have as few missing values as possible.

XV. What is the difference between prognostica and other software companies?

prognostica sees itself as the tailor-made "software solution for predictive analytics": Our forecasting solutions are individually adjusted to your needs. They are highly automated and correspond to the needs and specifications of your system, so you have access to unbiased and objective forecasts. Furthermore, we enable you to systematically process exogenous information.