This preview shows page 1. Sign up to view the full content.
The F statistic
and its corresponding pvalue can let us know how well this combination of
predictors predicts y.
The t statistics
and their corresponding pvalues can let us know how well
each predictor predicts
y
.
R
2
tells us how much of the variation in
y
is explained by variation in
the
x
scores
Residuals
Useful to determine how well the model 'fits'. The difference between the
y
value and
the predicted
y
value (error):
yŷ
. Calculate the residuals by predicting
y
for the actual
y
values
we used in the regression analysis. Then, look to see how well the model predicted the actual
data.
Multiple Regression Output

Seven
things to look for: 1)The equation of the regression model;
2)The ANOVA table with the F value for the overall test of the model; 3)The tratios, which test
the significance of the regression coefficients; 4)The value of SSEsum of squares of error.
Found under ANOVA section of regression table (The sum of all the residuals equal zero); 5)The
value of s
e
; 6)The value of R
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 01/26/2012 for the course QMST 2333 taught by Professor Mendez during the Spring '08 term at Texas State.
 Spring '08
 Mendez

Click to edit the document details