{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

qmst notes9

# qmst notes9 - Building Multiple Regression Models...

This preview shows pages 1–2. Sign up to view the full content.

Building Multiple Regression Models - Regression models -We are trying to predict unknown y values using known x values. To do so, we ' model' the data, or 'build a model' that gives us the best prediction based on the data we have. Standard Error of the Estimate (s e ) -A measure of how much error is in the regression model. It is preferable to have a small s e . Excel calls it just 'Standard Error' . Need to have a lot of content knowledge of what is being estimated to fully use the s e . Interaction -Occurs when the effects of one treatment ( x -variable) vary according to the levels of treatment of the other effect (used in medical comparisons). Ex : The interaction of temperature and humidity on a manufacturing process. Amount of red meat a family consumes based on religion. Linear regression - Predicting that scores will fall along a straight line. The simple regression and multiple regression examples we've done so far involved linear regression. p-value -Let's us know if the x variable is an accurate predictor of the y variable. R 2 - The proportion of variance of variance in the criterion variable (y) that is attributable to changes in the predictor variables. First order models -When the highest order of any predictor variable is 1 (i.e., x is not squared, cubed, etc.). Simple regression -First order model with one independent variable. Formula: ŷ = β 0 + β 1 x . Multiple Regression

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}