Prediction of financial (Profit & Loss) positions for different customers

We use internal data of various customers (currently TDG-Germany, EU, Group Development) to forecast different KPIs (key performance indicators) of customer choice. Currently, we use only univariate time series models, with the possible implementation of a multivariate approach in the future. The goal is to create as accurate KPIs predictions as possible to improve objectivity and efficiency in business forecasting and planning.

Technologies used:

  • Classic univariate TS models
  • Basic ones (mean, random walk)
  • Trend, seasonal, residual models (eg. RMA, SMA, SES)
  • Combined models (eg. ARIMA)