Využití prediktivní analýzy dat při predikci Cash Flow a její začlenění do podnikového reportingu

The main goal of this thesis is to confirm the hypothesis that predictive analysis and predictive models based on machine learning can by successfully used to predict corporate cash flow and how predicted values can be additionally incorporated into corporate reporting solution. The first of the partial goals is to use RStudio application and programming language R to create a set of prediction models. These models are evaluated, and the best machine learning technique is selected based on the lowest model prediction error. The second partial goal is to find and present the way, how to publish the best performing model and include the predicted values into the corporate reporting solution. Third partial goal is to repeat previous process using data science and machine learning platform, specifically Azure Machine Learning from Microsoft. The conclusion presents comparison of these two different approaches. The result of the work is a functional prototype of a solution for cash flow prediction which is incorporating the prediction values into corporate reporting solution.