The correct answer is: B. Test error is non zero.
A linear regression model is a statistical method that fits a linear equation to a set of data points. The model can be used to predict the value of a dependent variable (y) given the values of one or more independent variables (x).
The train error is the error that is made when the model is fit to the training data. The test error is the error that is made when the model is applied to the test data.
If the linear regression model perfectly fits the training data, then the train error will be zero. However, this does not mean that the test error will also be zero. The test error can be non zero because the model may not generalize well to the test data.
There are a number of reasons why the test error can be non zero even if the train error is zero. One reason is that the training data may not be representative of the test data. Another reason is that the model may be overfitting the training data. Overfitting occurs when the model learns the noise in the training data instead of the underlying relationship between the independent and dependent variables.
To reduce the test error, you can try to collect more training data, use a different model, or use a different training algorithm. You can also try to reduce the noise in the training data.
Here is a brief explanation of each option:
- Option A: Test error is also always zero. This is not always the case. The test error can be non zero even if the train error is zero.
- Option B: Test error is non zero. This is the correct answer. The test error can be non zero even if the train error is zero.
- Option C: Couldn’t comment on Test error. This is not a good option. You should be able to comment on the test error, even if you don’t know what it is.
- Option D: Test error is equal to Train error. This is not always the case. The test error can be different from the train error, even if the model perfectly fits the training data.