In this post, you will quickly learn about the **difference **between **predictive analytics **and **prescriptive analytics. **As data analytics stakeholders, one must get a good understanding of these concepts in order to decide when to apply predictive and when to make use of prescriptive analytics in analytics solutions / applications.

Without further ado, let’s get straight to the diagram.

In the above diagram, you could observe / learn the following:

**Predictive analytics**: In predictive analytics, the model is trained using historical / past data based on supervised, unsupervised, reinforcement learning algorithms. Once trained, the new data / observation is input to the trained model. The output of the model is prediction in form of regression (numerical estimate), classification (binary or multi-class classification), clusters (segmenting the data in groups based on similarity) etc.**Prescriptive analytics**: In prescriptive analytics, one or more mathematical algorithms are applied on the outcomes of predictive analytics solutions / predictions (optional) and business goals, and, the best solution is recommended. The recommended solution optimises the business goal, taking into consideration all decision variables, constraints, and trade-offs. Prescriptive analytics is primarily related to**decision optimization**problems.**Prescriptive analytics builds upon the results of predictive analytics.**It suggests all favourable outcomes and, which courses of action needs to be taken to reach a particular outcome. Recommendation systems are classical examples where prescriptive analytics is applied. Here are some examples of prescriptive analytics solutions:**Google’s self-driving car**: Decision on when and where to turn, whether to slow down or speed up and when to change lanes is done using prescriptive analytics methods.- In
**sourcing**, the factors affecting pricing is considered to get the best terms and appropriately hedge risks.

Here is another diagram which depicts the path from descriptive analytics to prescriptive analytics. Quite a self explanatory diagram. Note the **difference between predictive and prescriptive analytics**.

## References

Here are some good links to understand the concepts of **predictive **and **prescriptive analytics:**

- Generative Modeling in Machine Learning: Examples - March 19, 2023
- Data Analytics Training Program (Beginners) - March 18, 2023
- Histogram Plots using Matplotlib & Pandas: Python - March 18, 2023

[…] Predictive & prescriptive analytics: In predictive analytics, the model is trained using historical / past data based on supervised, unsupervised, reinforcement learning algorithms. Once trained, the new data / observation is input to the trained model. The output of the model is the prediction (what will happen in future) in form of regression (numerical estimate), classification (binary or multi-class classification), clusters (segmenting the data in groups based on similarity) etc. In prescriptive analytics, one or more mathematical algorithms are applied on the outcomes of predictive analytics predictions and business goals, and, the best solution is recommended. You may want to check out this post to read further details – Difference between predictive and prescriptive analytics […]