Prediction
Outline
Topics
- Prediction using decision trees
- Example
Rationale
Often we do not care so much about “parameters” but instead about predicting future observations.
Example: coins in a bag
Consider the setup from last week with 3 coins and 3 flips with
Question: given you have see 3 heads, what is the probability that the next one is also heads?
Mathematically:
General approach
Key message: In Bayesian statistics, prediction and parameter estimation are treated in the exact same way!
Idea: Add
Then, to compute
Example, continued
Use the following picture to help you computing
Notation: let
Question: compute
- None of the above
- There is only one way to get
: this has to be the standard coin, i.e., - To compute the probability of that path we can multiply the edge probabilities (why?):
Question: compute
- None of the above
- Twist: two distinct paths are compatible with the event:
- Sum the probabilities of the paths leading to the same prediction (why can we do this?).:
Question: compute the predictive
- None of the above
Let:
Note:
Hence: