Statistical modeling is about understanding contexts and making forecasts where coincidences and uncertainty are involved.
With the use of statistical modeling, underlying knowledge about a process is connected with data, in an objective way. The statistical models can contain a number of unknown parameters and are thus flexible. The parameters are quantified from data using different techniques. Decisions are thus based on facts and the best possible balance between the probabilities of different outcomes.
The terms statistical modeling and machine learning are often used interchangeably, but machine learning finds patterns in data to a greater extent without the use of models. NR’s great strength is that we have extensive experience and broad knowledge of this entire toolbox. We customize and find the right method for each problem. This gives the most precise and realistic forecasts and causal relationships, including an understanding of the uncertainty in the answers when it is important.