A time series is a sequence of observations taken sequentially over time. SEDGE makes extrapolations or forecasts possible by one or multiple targets by taking models built on historical data and using this to predict future observations against a time component.
Time series is used in many areas such as:
Sales and budget forecasting
Stock market analysis
Inventory studios and many more
SEDGE’s time series forecasting functionality allows data analysts to forecast what is going to happen in the future based on trends, seasonality, cyclicity and irregularity, with greater accuracy and efficiency. SEDGE’s time series capabilities solve the many prediction problems typically associated with a time component.
Our step-by-step Time Series Forecasting Process
To get to this prediction, SEDGE provides the following data analysis functionalities.
The data set is profiled, with options available for data slicing to allocate a sample of data in a certain period of time, as well as data protection.
Time Series modeling
Univariate or Multivariate models forecast future data sets over time based on trend, seasonality and irregularities.
The SEDGE application now allows data analysts to review and compare models by MAPE (mean absolute percentage error), RMSE (root mean square error), MSE (mean squared error), MAE (mean absolute error), and MLCH. This allows the analyst to save and test models for greatest accuracy.