FRM Part I · FRM Exam Part I · Machine-Learning Methods
An analyst standardizes a feature using the training sample mean of 12 and a training standard deviation of 4. A new observation in the test set has a raw value of 20. What is its standardized value, and what is the correct procedure regarding the parameters used?
The standardized value is 2.0, calculated as (20 - 12) / 4 using the training mean and standard deviation. Test data should be scaled with parameters from the training sample, not its own statistics, to prevent data leakage and keep the transformation consistent between training and testing.
- A2.0, using the training-sample mean and standard deviationCorrect
- B2.0, after recomputing the mean and standard deviation on the test set
- C8.0, using the training-sample mean and standard deviation
- D0.5, using the training-sample mean and standard deviation
Explanation
z = (20 - 12) / 4 = 2.0. Test data should be transformed with the parameters estimated from the training data to avoid data leakage and keep the transformation consistent. Option 8.0 omits division by the standard deviation, and 0.5 divides by the wrong quantity (20 - 12 divided by 16, the variance).
Did you get it right without looking?
One question tells you little. A timed set on Machine-Learning Methods shows your real accuracy, how long you take and where you lose marks.
More Machine-Learning Methods questions
- An analyst has 10,000 observations to forecast corporate bond downgrades. She first standardizes all features using the mean and standard de…
- A logistic regression for default has the form ln(p/(1-p)) = -3.0 + 0.5 x, where x is the debt-to-income ratio expressed in percentage point…
- A risk analyst fits a single unpruned decision tree to predict loan default and finds that it classifies the training data almost perfectly …
- Which statement best distinguishes a random forest from simple bagging of decision trees?
- A neuron has two inputs x1 = 2 and x2 = -1, weights w1 = 0.5 and w2 = 1.5, and bias b = 0.25. The neuron uses a ReLU activation, f(z) = max(…
- A risk team chooses the penalty parameter lambda for an elastic net model used to predict loan defaults. Which procedure is most appropriate…