Skip to Main Content

Applied Regression Analysis by John Fox Chapter 3: Examining Data | Stata Textbook Examples

 Page 34, figure 3.1.

use https://stats.idre.ucla.edu/stat/stata/examples/ara/quartet, clear

graph twoway (scatter y1 x1) (lfit y1 x1), xlabel(0(5)20) ylabel(0(5)15)
Image ara3_1
graph twoway (scatter y2 x2) (lfit y2 x2), xlabel(0(5)20) ylabel(0(5)15)
Image ara3_2
graph twoway (scatter y3 x3) (lfit y3 x3), xlabel(0(5)20) ylabel(0(5)15)
Image ara3_3
graph twoway (scatter y4 x4) (lfit y4 x4), xlabel(0(5)20) ylabel(0(5)15)
Image ara3_4

Page 37, figure 3.2.  We have skipped this for now.

Page 38, figure 3.3 This is a stem and leaf plot. The plot looks a bit different from
Fox due to a different style of display. Also, Stata rounded to the nearest hundred, but Fox truncated the data.
For example, 1890 was displayed at 1|8 in Fox, but 1|9 in Stata.

use https://stats.idre.ucla.edu/stat/stata/examples/ara/prestige, clear

stem income, round(100)
Stem-and-leaf plot for income (Average income, dollars)

income rounded to nearest multiple of 100
plot in units of 100

   0* | 69
   1* | 79
   2* | 44689
   3* | 001125556667999
   4* | 01223345677788
   5* | 1111234566889
   6* | 012335566799
   7* | 01145679
   8* | 0000123344889999
   9* | 36
  10* | 4
  11* | 004
  12* | 45
  13* | 
  14* | 026
  15* | 
  16* | 
  17* | 5
  18* | 
  19* | 3
  20* | 
  21* | 
  22* | 
  23* | 
  24* | 
  25* | 39

Page 40, figure 3.4.  We have skipped this for now.

Page 41, figure 3.5 This figure shows an example of a kernel density estimator, which can be obtained
in Stata with the kdensity command. The width(800) option is used to
specify the half-width of 800.

kdensity income, xlabel(0(5000)30000) ylabel(0(.00005).00015) width(800)
Image ara3_5

Page 47, figure 3.11 You can get a boxplot in Stata using the graph
box
command.

graph box income, ylabel(0(10000)30000)
Image ara3_6

Page 51, figure 3.12 We can use the graph twoway command with two variables to obtain a scatterplot like figure 3.12.

use https://stats.idre.ucla.edu/stat/stata/examples/ara/vocab, clear

graph twoway (scatter vocab educ) (lfit vocab educ), ylabel(0 5 10)
Image ara3_7

Page 51, figure 3.13. We can use the jitter(5) option to jitter the points and
provide a more readable scatterplot.

graph twoway (scatter vocab educ, jitter(5)) (lfit vocab educ), ylabel(0 5 10)
Image ara3_8

Page 52, figure 3.14 You can make side by side boxplots for the same variable in Stata by using
the over(nation) option on the graph command.

use https://stats.idre.ucla.edu/stat/stata/examples/ara/ornstein, clear

graph box intrlcks, over(nation)
Image ara3_9

Page 54, figure 3.15 shows a scatterplot matrix using the graph matrix
command.

use https://stats.idre.ucla.edu/stat/stata/examples/ara/prestige, clear

graph matrix prestige income educat
Image ara3_10

Page 55, figure 3.16.

use https://stats.idre.ucla.edu/stat/stata/examples/ara/davis, clear

graph twoway (scatter measwt reptwt if female == 1, msymbol(Oh) yvarlabel("female")) ///
	(scatter measwt reptwt if female == 0, msymbol(t) yvarlabel("male")) ///
	(function y = x, range(40 140) yvarlabel("y = x")), ylabel(0(50)200) legend(rows(1))
Image ara3_11

Page 56, figure 3.17. Unfortunately, Stata does not have 3-D graphs.