What does R-squared tell you about a regression model?
Why D is the answer
How much variance it explains
It is the proportion of explained variance.
In detail
- R-squared is the proportion of explained variance.
- It does not prove cause.
Use case
An illustrative scenario, not a real incident.
An R-squared of 0.8 means that the model explains 80% of the variance.
See it in different ways
mindmap
root((Data analysis))
Metrics
Loss
R-squared
Charts
Line: over time
Scatter: two variables
Care
Same data
Confounding factorsmindmap
root(("R-squared"))
n1["Near 1"]
n2["Explains most variance"]
n3["Near 0"]
n4["Explains little"]Key point: R-squared is explained variance.
Why the other options are wrong
- A. How many parameters the model contains in total
- That is a parameter count.
- B. How long the model took to complete its training
- That is training time.
- C. How many rows of data were removed before training
- That is data cleaning.