The difference between the observed value and the model-predicted value is known as what?

The difference between the observed value and the model-predicted value is known as what?

• A. Deviance
• B. Residual
• C. Error
• D. Bias

Answer: (A)

The difference for a single case between what you observe and what your model predicts is called the residual. It’s computed as observed minus predicted and shows how far off the model is for that data point. Analyzing residuals helps you assess model fit—small, randomly scattered residuals suggest a good fit, while large or patterned residuals hint at problems like nonlinearity or heteroscedasticity. Deviance is a overall fit measure used in certain models, aggregating discrepancies across all observations, not the per-case difference. Error is a general term for the true, unobserved deviation, while residuals are the observable estimates of that deviation for your data. Bias refers to systematic error in an estimator, not the individual discrepancy between observed and predicted values.

The difference for a single case between what you observe and what your model predicts is called the residual. It’s computed as observed minus predicted and shows how far off the model is for that data point. Analyzing residuals helps you assess model fit—small, randomly scattered residuals suggest a good fit, while large or patterned residuals hint at problems like nonlinearity or heteroscedasticity. Deviance is a overall fit measure used in certain models, aggregating discrepancies across all observations, not the per-case difference. Error is a general term for the true, unobserved deviation, while residuals are the observable estimates of that deviation for your data. Bias refers to systematic error in an estimator, not the individual discrepancy between observed and predicted values.