click below
click below
Normal Size Small Size show me how
WGU-748
SAS Programming II Business Analysis Applications
| Question | Answer |
|---|---|
| VIF Variance Inflation Factor | Detects Multi-collinearity 1/(1-R-square) As a rule of thumb, should not exceed 10 |
| -2 Log L | Likelihood-ratio statistic Larger values lead to smaller p-values |
| Concordant | Observations with a larger value of X also have a larger value of Y |
| Discordant | Observations with a larger value of X have a smaller value of Y |
| odds ratio | odds1/odds0 |
| /solution | Option used in proc GLM for models with classification effects |
| ilink | requests logit estimate be transformed into probabilities |
| Type I error | Power is decreased As alpha is decreased Type I error rejecting null hypothesis when it is true (false pos) |
| Type II error | Failing to reject the null hypothesis when it is false (false neg) |
| Standardized Estimates (e.g. "What is the rank order of the relative importance of the predictor variables?") | variables are rank-ordered by their absolute values |
| Which statistic, calculated from a validation sample, can help decide which model to use for prediction of a binary target variable? | Average squared error |