Predictive analytics
What is it?
Predictive analytics encompasses a variety of statistical techniques from data mining, predictive modelling, and machine learning that analyze current and historical facts to make predictions about future or otherwise unknown events.
Do you need it?
Predictions are extremely important in our private and business life. Like we plan weekend activities based on the weather forecast, we can plan resources based on sales forecast. It saves time and money. However past data is needed, the more the better.
How do we do it?
1. Data analysis
It is a prerequisite for accurate predictions.
2. Find best indicators
Some data features are good indicators for predictions, others are irrelevant. Statistical and machine learning methods help us to select the best only.
3. Select appropriate model
There are a large variety of models, some of them are simple to understand, others are extremely complex. We always train multiple and select a few best performers.
4. Evaluate performance
Model performance must be evaluated in order to be trusted. We always check how model performs on the unseen data.
5. Present results
It is important to understand where the predicted number came from. In addition to the actual predictions, we provide behind the scenes model informations. Sometimes those are even more important since we can learn new relationships in the data.