A note on computing with SAS (Version 9):
The SAS GENMOD procedure is used for fitting the Poisson regression models of
Chapter 12. This procedure is very general. It can also be used for the logistic
regression models in Chapter 11, as well as
6.1 (a) The Minitab output of the three regressions is shown below.
In the model involving x 1 alone, the hypothesis 1 = 0 can not be rejected. This
indicates that x 1 by itself is not important.
Similarly, in the model involving x 2 alone, x 2
01 - Midterm Test
Statistics 3859/9859A. October 2, 2014.
7:00pm to 9:00pm
Talbot College (TC) Rms. 341 and 342
Copyright A. I. McLeod, 2014
Bring to the Exam
1. Watch (no cell phones, ipads, tablets or laptops allowed!)
2. Your student ID card
A note on computing in time series situations
The Minitab software is used here for calculating the autocorrelation function of time
series observations and for fitting the autoregressive integrated moving average
(ARIMA) models in Chapter 10.
A note on computing with SPSS (Version 11.5):
We use the SPSS software to fit the nonlinear regression models of Chapter 9. SPSS
works like a spreadsheet program. We enter the data into the various columns of the
spreadsheet and use the tabs: An
(a) Backward elimination: Drop x 3 (step 1); drop x 4 (step 2); next candidate x 2 for
elimination can not be dropped. Model with x 1 and x 2 .
(b) Forward selection: Enter x 4 (step1); enter x 1 (step 2); enter x 2 (step 3); next
1.1 Tensile strength of an alloy can be expected to increase with increasing hardness and
density of the stock. Bivariate scatter plots of tensile strength against hardness and of
tensile strength against density of the stock are useful. Such sc
(a) The Minitab output of various regression models is given below. For each fitted
model we list the estimated equation (with estimates, standard errors, and p-values),
the coefficient of determination R2, the root mean square error s, and
5.1 Interaction; bonus for having a MBA; furthermore, salary increases faster for
5.2 (a) $ 3,000; (b) $ 900
Body Fat (%)
Many excellent computer programs are available for plotting the data and for carrying
out the regression calculations. Here we use S-Plus, R, Minitab, SAS, and SPSS. Most
programs work the same and it is not difficult to switch from one program