algotrading101 / conformal-prediction-mapie-algotrading101

Conformal Prediction - A Practical Guide with MAPIE

Conformal-Prediction-MAPIE-Algotrading101

Conformal Prediction - A Practical Guide with MAPIE

Conformal Prediction is, in essence, a method for constructing prediction intervals with a guaranteed level of accuracy, regardless of the distribution of the underlying data.

One of the most important advantages of Conformal Prediction is that it allows for the inclusion of uncertainty in predictions. In many traditional machine learning techniques, we often try to optimize for the highest possible accuracy or any other metric. While this may be effective in some cases, it can also lead to overconfidence in our predictions, especially when dealing with complex or noisy data.

Conformal Prediction addresses this issue by providing a way to quantify and communicate the uncertainty in our predictions. This is especially useful in industries where the cost of a wrong prediction is high, such as finance or healthcare.

Table of contents

1. What is Conformal Prediction?
2. What is Conformal Prediction used for?
3. Why should I use Conformal Prediction?
4. Why shouldn’t I use Conformal Prediction?
5. How can Conformal Prediction be used in Finance?
6. How can Conformal Prediction be used in Algorithmic Trading?
7. What are some Conformal Prediction alternatives?
8. Understanding Conformal Prediction
9. What is MAPIE?
10. How to get started with Conformal Prediction?
11. How to apply Conformal Prediction with MAPIE in Python?
12. How to apply Conformal Prediction for Classification with MAPIE?
13. How to apply Conformal Prediction for Regression with MAPIE?
14. How to perform forecasting with Conformal Prediction in Python?
15. Where can I learn more about Conformal Prediction?
16. Full code

---

Info

Author	Igor Radovanovic
Published	December 20, 2022

conformal-prediction-mapie-algotrading101's Issues

question - Does MAPIE support multiple time series in tabular format?

I wonder whether MapieTimeSeriesRegressor supports building prediction interval when input data is tabular form and contains multiple time series. Specifically how does it know to pull the right residuals in BlockBootstrap to build distribution when the scales of times series in input data are significantly different?