Which statement about DFBeta is true?

Which statement about DFBeta is true?

• A. A standardized version exists; a common rule uses absolute values greater than 2 as a cutoff.
• B. It measures the global fit of the model.
• C. It is the same as DFFit.
• D. It is used to detect multicollinearity.

Answer: (A)

DFBETA tells you how much a single observation changes each regression coefficient when that observation is removed. For every case and each predictor, it shows the size of the impact on the corresponding beta estimate if that case were deleted. There is a standardized version of these values, which puts the effects on a common scale so you can compare influence across different coefficients. A commonly used rule of thumb is that if the absolute value of a standardized DFBETA exceeds 2, that observation is considered influential for that specific coefficient. This helps you identify data points that unduly affect the estimated relationships, rather than indicating a problem with the overall model fit or with multicollinearity.

DFBETA tells you how much a single observation changes each regression coefficient when that observation is removed. For every case and each predictor, it shows the size of the impact on the corresponding beta estimate if that case were deleted. There is a standardized version of these values, which puts the effects on a common scale so you can compare influence across different coefficients. A commonly used rule of thumb is that if the absolute value of a standardized DFBETA exceeds 2, that observation is considered influential for that specific coefficient. This helps you identify data points that unduly affect the estimated relationships, rather than indicating a problem with the overall model fit or with multicollinearity.