SOA Exam SRM Sample Question 76
Study guide: Exam SRM
You have run a principal components analysis on a veterinarian's database of dogs. The result is provided below.
| PC1 | PC2 | PC3 | PC4 | PC5 | PC6 | PC7 | PC8 | PC9 | PC10 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Weight (lb) | 0.506 | 0.017 | 0.003 | -0.002 | 0.004 | -0.002 | 0.006 | -0.072 | -0.389 | -0.766 |
| Height (in) | 0.504 | 0.009 | 0.003 | -0.004 | 0.004 | -0.003 | -0.002 | -0.104 | -0.575 | 0.636 |
| Food (cups per day) | 0.496 | 0.013 | -0.006 | -0.002 | 0.004 | 0.010 | 0.018 | -0.603 | 0.620 | 0.070 |
| Stride Length (inches) | 0.493 | 0.015 | 0.002 | 0.001 | -0.006 | -0.010 | -0.005 | 0.787 | 0.364 | 0.067 |
| Age (years) | -0.007 | 0.275 | -0.628 | 0.066 | 0.469 | -0.493 | -0.252 | -0.004 | 0.002 | 0.000 |
| Vet Visits per Year | -0.002 | 0.197 | -0.648 | 0.036 | -0.680 | 0.278 | 0.031 | 0.001 | -0.008 | 0.001 |
| Miles Walked per Week | 0.009 | -0.566 | -0.239 | 0.007 | -0.080 | -0.424 | 0.661 | 0.006 | -0.004 | 0.000 |
| Resting HR | -0.014 | 0.461 | 0.350 | -0.040 | -0.471 | -0.663 | 0.039 | -0.014 | 0.002 | 0.002 |
| Body Fat (%) | -0.021 | 0.593 | 0.009 | 0.027 | 0.297 | 0.246 | 0.705 | 0.015 | -0.007 | 0.007 |
| Barks per Day | 0.004 | -0.019 | 0.081 | 0.996 | -0.033 | -0.008 | -0.006 | -0.003 | -0.002 | 0.001 |
Determine which of the following statements is true.
- A) Loadings close to 0 represent noise in the dataset.
- B) PCA is only capturing the linear relationships between the predictors.
- C) Principal components determine features contributing the least amount of variance.
- D) The loading of PC1 for Age (years) indicates that it is not a strong predictor and can be discarded.
- E) The first several principal components are capturing the noise and should not be used in a predictive model.
Answer this question and see the worked solution