CAS Exam MAS-II Fall 2018, Question 36
Study guide: Exam MAS-II
A training data set contains eight observations for two predictor variables, X_1 and X_2, and a response variable, Y. The response Y has three possible classes P, N, and U.
| i | X_1 |
X_2 |
Y | Distance from (x_{1i}, x_{2i}) to (3, 2) |
|---|---|---|---|---|
| 1 | 4.1 | 3.0 | P | 1.5 |
| 2 | -2.6 | -3.0 | N | 7.5 |
| 3 | -1.1 | 1.3 | U | 4.2 |
| 4 | 0.0 | 1.2 | U | 3.1 |
| 5 | -3.0 | -5.0 | N | 9.2 |
| 6 | 2.0 | 2.0 | U | 1.0 |
| 7 | -3.1 | -2.0 | N | 7.3 |
| 8 | 3.2 | 3.1 | P | 1.1 |
Three models are constructed using K-Nearest Neighbors and the data set above to predict Y for the two-dimensional space of predictors X_1 and X_2.
- Model I: K-Nearest Neighbors with K=1.
- Model II: K-Nearest Neighbors with K=3.
- Model III: K-Nearest Neighbors with K=7.
Each model is used to classify the point (3, 2).
Determine the predicted response Y at this point using each of the three models.
- A) Model I: Y = P, Model II: Y = P, Model III: Y = P
- B) Model I: Y = P, Model II: Y = P, Model III: Y = U
- C) Model I: Y = U, Model II: Y = P, Model III: Y = P
- D) Model I: Y = U, Model II: Y = P, Model III: Y = U
- E) The answer is not given by (A), (B), (C), or (D).
Answer this question and see the worked solution