20. Measuring and Modeling Relationships between Variables Flashcards
Sample Correlation Coefficient (r) P.393
A statistics that measures the degree of linear relationship between two sets of numbers.
r= Sxy / √Sxx Syy
Coefficient of Determinations r^2 P.395
Amount of the variation in Y that is explained by the fitted simple linear equation.
Adjusted Coefficient of Determination r^2adj
Used with multiple regression. Adjusted coefficient will increase when variables are added to the model that act to decrease the mean square error.
Regression analysis P.397
A technique that typically uses continuous predictor variables to predict the variation in a continuous response variable.
Linear regression coefficients P.398
Numbers associated with each predictor variable in a linear regression equation that tells how the response variable changes with each unit increase in the predictor variable.
Simple Linear Regression P.398
Y=b0+b1X+E
Multicollinearity P.406
Two or more predictor variables in a multiple regression model are correlated.
Multivariate Tools P406
- Principal components
- Factor analysis
- Discriminant analysis
- Multiple ANOVA
Principal Components
Used to form a smaller number of uncorrelated variables from a large set of data. The goal is to explain the maximum amount of variance with the fewest number of principal components.
Factor analysis
Used to determine the underlying factors responsible for correlations in the data.
Discriminant analysis
Used to classify observation into two or more groups if you have a sample with known groups.
MANOVA
Used to analyze both balanced and unbalanced experimental designs.
Multi-vari P.417
Graphical technique for viewing multiple sources of process variation.
Positional variation P.417
Within-part variation
Cyclical variation P.417
Part-to-part variation
