Which term describes adjusting variables to have a mean of zero and a variance of one, especially when outliers are present?

Which term describes adjusting variables to have a mean of zero and a variance of one, especially when outliers are present?

• A. Standardization
• B. Normalization (Min-Max)
• C. Data Transformation
• D. Principal Components Analysis (PCA)

Answer: (A)

Standardization is the process of adjusting a variable by subtracting its mean and dividing by its standard deviation, so the transformed values have a mean of zero and a variance of one. This puts features on a common scale, which helps many algorithms—especially those that rely on distances or assume inputs are centered and similarly scaled—work more effectively. When outliers are present, the mean and standard deviation are influenced, but standardization remains advantageous over min–max normalization. Min–max uses the extreme values (minimum and maximum) to set the range, so outliers can distort the scale and squeeze the bulk of the data into a narrow interval. By centering at zero and scaling by the spread, standardized data preserve the relative structure of the majority of observations, even with some extreme values. PCA also benefits from standardized data because the methods rely on the covariance structure, which is sensitive to scale. In short, standardization explicitly creates a zero mean and unit variance, making features comparable and suitable for many analyses.

Standardization is the process of adjusting a variable by subtracting its mean and dividing by its standard deviation, so the transformed values have a mean of zero and a variance of one. This puts features on a common scale, which helps many algorithms—especially those that rely on distances or assume inputs are centered and similarly scaled—work more effectively.

When outliers are present, the mean and standard deviation are influenced, but standardization remains advantageous over min–max normalization. Min–max uses the extreme values (minimum and maximum) to set the range, so outliers can distort the scale and squeeze the bulk of the data into a narrow interval. By centering at zero and scaling by the spread, standardized data preserve the relative structure of the majority of observations, even with some extreme values.

PCA also benefits from standardized data because the methods rely on the covariance structure, which is sensitive to scale. In short, standardization explicitly creates a zero mean and unit variance, making features comparable and suitable for many analyses.