Term
|
Definition
| analysis of the equational relationship between X and Y. |
|
|
Term
|
Definition
| , has a t distribution which standardizes its value to see if it is significantly different from 0. |
|
|
Term
|
Definition
| , ß0, is not interpreted. |
|
|
Term
|
Definition
| ,r, indicates direction and strength of the LINEAR relationship between X and Y. |
|
|
Term
| Coefficient of Determination |
|
Definition
| R2, the ratio of explained variation in Y to the total variation. |
|
|
Term
|
Definition
| will be smaller for better predictive equations. |
|
|
Term
|
Definition
| is the study of the nature and degree of the relationship between variables. |
|
|
Term
|
Definition
| is using Xs beyond the range of the given Xs to predict Y. This can cause large errors in prediction. |
|
|
Term
| Relationship of slope to the correlation coefficient |
|
Definition
|
|
Term
|
Definition
| when Xs are highly correlated this gives redundant information. |
|
|
Term
|
Definition
| - non-constant variance in the residuals |
|
|
Term
|
Definition
| atypical values in a data set. |
|
|
Term
| Non-linear relationships- |
|
Definition
| curvilinear patterns or logarithmic relationships |
|
|
Term
| Multiple regression analysis |
|
Definition
| includes one dependent variable and more than one independent. |
|
|
Term
|
Definition
| tries all combinations of variables and produces the best predictors in order of their predictive power. |
|
|
Term
| Artificially Inflated R-squared |
|
Definition
| occurs when there are too many predictors and not enough samples. |
|
|
Term
|
Definition
| should produce a nearly straight line without outliers. |
|
|
Term
| T distribution vs. F distribution |
|
Definition
| : t is used to test the individual coefficients where F tests the overall model. |
|
|
Term
| Residuals are the differences in the observed value of Y at a given X and the predicted value. |
|
Definition
|
|
Term
|
Definition
| should fall within +/- 3 in order to be considered normal values. |
|
|
