Welcome to the Institute for Digital Reseach and Education
Institute for Digital Reseach and Education Home
Help the Stat Consulting Group by giving a gift
Stata Textbook Examples
Applied Regression Analysis by John Fox
Chapter 8: Analysis of Variance
Table in the middle of page 160 on data file duncan.
use http://www.ats.ucla.edu/stat/stata/examples/ara/duncan, clear (From Fox, Applied Regression Analysis. Use 'notes' command for source of data) sort occ_type by occ_type: summarize prestige -> occ_type= bc Variable | Obs Mean Std. Dev. Min Max ---------+----------------------------------------------------- prestige | 21 22.7619 18.05521 3 67 -> occ_type= prof Variable | Obs Mean Std. Dev. Min Max ---------+----------------------------------------------------- prestige | 18 80.44444 14.10558 45 97 -> occ_type= wc Variable | Obs Mean Std. Dev. Min Max ---------+----------------------------------------------------- prestige | 6 36.66667 11.79265 16 52
Figure 8.1, page 161 using teh data file duncan.
graph box prestige, over(occ_type) ylabel(0 50 100)
Table in the middle of page 161 using the data file duncan.
anova prestige occ_type
Number of obs = 45 R-squared = 0.7574
Root MSE = 15.8847 Adj R-squared = 0.7459
Source | Partial SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 33090.0571 2 16545.0286 65.57 0.0000
|
occ_type | 33090.0571 2 16545.0286 65.57 0.0000
|
Residual | 10597.5873 42 252.323507
-----------+----------------------------------------------------
Total | 43687.6444 44 992.90101
Table 8.2 on page 167 using the data file moore.
use http://www.ats.ucla.edu/stat/stata/examples/ara/moore, clear
(From Fox, Applied Regression Analysis. Use 'notes' command for source of data)
table status fcat, contents(n conform mean conform sd conform)
----------+-----------------------------
Status of | F-scale categorized
partner | high low medium
----------+-----------------------------
high | 7 5 11
| 11.85714 17.4 14.27273
| 3.933979 4.505552 3.951985
|
low | 8 10 4
| 12.625 8.9 7.25
| 7.347254 2.643651 3.947573
----------+-----------------------------
Figure 8.5 on page 169 using the data file moore. In the first part, we use a anovaplot program. Next, we will try to create the graph directly.
Using anovaplot program. You can download anovaplot from within Stata by typing findit anovaplot (see How can I use the findit command to search for programs and get additional help? for more information about using findit).
recode fcat 1=3 2=1 3=2 anovaplot, scatter(msymbol(none)) ylabel(5(5)20)
Next we do it directly.
use http://www.ats.ucla.edu/stat/stata/examples/ara/moore, clear (From Fox, Applied Regression Analysis. Use 'notes' command for source of data) recode fcat 1=3 2=1 3=2 label define flab 1 low 2 medium 3 high label value fcat flab egen xmeanh = mean(conform) if (status==1), by(fcat) egen xmeanl = mean(conform) if (status==2), by(fcat) graph twoway (scatter xmeanl fcat, connect(l) sort) /// (scatter xmeanh fcat, connect(l) sort), xlabel(1 2 3) ylabel(5(5)20)
Figure 8.6 on page 170 using the data file moore.
graph twoway (scatter conform fcat if status ==1, jitter(5)) /// (scatter xmeanh fcat, connect(l) sort), xlabel(1 2 3) ylabel(5 15 25)
graph twoway (scatter conform fcat if status ==2, jitter(5)) /// (scatter xmeanl fcat, connect(l) sort), xlabel(1 2 3) ylabel(5 15 25)
Results on page 177 using the data file moore.
use http://www.ats.ucla.edu/stat/stata/examples/ara/moore, clear (From Fox, Applied Regression Analysis. Use 'notes' command for source of data ) gen c1=1 if (fcat==1) gen c2=0 if (fcat==1) replace c1=0 if (fcat==2) replace c2=1 if (fcat==2) replace c1=-1 if (fcat==3) replace c2=-1 if (fcat==3) gen r=1 if(status==1) replace r=-1 if(status==2) gen rc1=r*c1 gen rc2=r*c2
The anova procedures below give the sum of squares on page 177 and the tests yield table 8.6 on page 178. Also notice that the F-values for the case alpha|beta and the case beta|alpha are different from the results in the book as different degree of freedom has been used in both cases.
anova conform r c1 c2 rc1 rc2 , se cont(r c1 c2 rc1 rc2)
Number of obs = 45 R-squared = 0.3237
Root MSE = 4.57912 Adj R-squared = 0.2370
Source | Seq. SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 391.436039 5 78.2872078 3.73 0.0074
|
r | 204.332411 1 204.332411 9.74 0.0034
c1 | 7.92747828 1 7.92747828 0.38 0.5422
c2 | 3.68722176 1 3.68722176 0.18 0.6773
rc1 | 111.656569 1 111.656569 5.33 0.0264
rc2 | 63.8323592 1 63.8323592 3.04 0.0889
|
Residual | 817.763961 39 20.9683067
-----------+----------------------------------------------------
Total | 1209.20 44 27.4818182
test r
Source | Partial SS df MS F Prob > F
-----------+----------------------------------------------------
r | 239.56237 1 239.56237 11.42 0.0017
Residual | 817.763961 39 20.9683067
test rc1 rc2
Source | Partial SS df MS F Prob > F
-----------+----------------------------------------------------
rc1 rc2 | 175.488928 2 87.7444639 4.18 0.0226
Residual | 817.763961 39 20.9683067
test c1 c2
Source | Partial SS df MS F Prob > F
-----------+----------------------------------------------------
c1 c2 | 36.0187056 2 18.0093528 0.86 0.4315
Residual | 817.763961 39 20.9683067
anova conform r c1 c2, se cont(r c1 c2)
Number of obs = 45 R-squared = 0.1786
Root MSE = 4.92196 Adj R-squared = 0.1185
Source | Seq. SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 215.947111 3 71.9823704 2.97 0.0428
|
r | 204.332411 1 204.332411 8.43 0.0059
c1 | 7.92747828 1 7.92747828 0.33 0.5704
c2 | 3.68722176 1 3.68722176 0.15 0.6985
|
Residual | 993.252889 41 24.2256802
-----------+----------------------------------------------------
Total | 1209.20 44 27.4818182
test r
Source | Partial SS df MS F Prob > F
-----------+----------------------------------------------------
r | 212.213778 1 212.213778 8.76 0.0051
Residual | 993.252889 41 24.2256802
test c1 c2
Source | Partial SS df MS F Prob > F
-----------+----------------------------------------------------
c1 c2 | 11.6147 2 5.80735002 0.24 0.7879
Residual | 993.252889 41 24.2256802
anova conform r rc1 rc2, se cont(r rc1 rc2)
Number of obs = 45 R-squared = 0.2939
Root MSE = 4.56333 Adj R-squared = 0.2423
Source | Seq. SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 355.417333 3 118.472444 5.69 0.0024
|
r | 204.332411 1 204.332411 9.81 0.0032
rc1 | 85.0926235 1 85.0926235 4.09 0.0498
rc2 | 65.9922988 1 65.9922988 3.17 0.0825
|
Residual | 853.782667 41 20.8239675
-----------+----------------------------------------------------
Total | 1209.20 44 27.4818182
anova conform c1 c2 rc1 rc2, se cont(c1 c2 rc1 rc2)
Number of obs = 45 R-squared = 0.1256
Root MSE = 5.14132 Adj R-squared = 0.0382
Source | Seq. SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 151.873669 4 37.9684173 1.44 0.2398
|
c1 | .133333333 1 .133333333 0.01 0.9437
c2 | 3.60 1 3.60 0.14 0.7140
rc1 | 82.6026667 1 82.6026667 3.12 0.0847
rc2 | 65.5376692 1 65.5376692 2.48 0.1232
|
Residual | 1057.32633 40 26.4331583
-----------+----------------------------------------------------
Total | 1209.20 44 27.4818182
anova conform r, se cont(r)
Number of obs = 45 R-squared = 0.1690
Root MSE = 4.83415 Adj R-squared = 0.1497
Source | Seq. SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 204.332411 1 204.332411 8.74 0.0050
|
r | 204.332411 1 204.332411 8.74 0.0050
|
Residual | 1004.86759 43 23.3690137
-----------+----------------------------------------------------
Total | 1209.20 44 27.4818182
anova conform c1 c2, se cont(c1 c2)
Number of obs = 45 R-squared = 0.0031
Root MSE = 5.35739 Adj R-squared = -0.0444
Source | Seq. SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 3.73333333 2 1.86666667 0.07 0.9371
|
c1 | .133333333 1 .133333333 0.00 0.9460
c2 | 3.60 1 3.60 0.13 0.7250
|
Residual | 1205.46667 42 28.7015873
-----------+----------------------------------------------------
Total | 1209.20 44 27.4818182
Result in the middle of page 192 using the data file moore.
use http://www.ats.ucla.edu/stat/stata/examples/ara/moore, clear
(From Fox, Applied Regression Analysis. Use 'notes' command for source of data> )
gen d=1 if(status==2)
(23 missing values generated)
replace d=0 if(status==1)
(23 real changes made)
gen intfd=fscore*d
reg conform fscore d intfd
Source | SS df MS Number of obs = 45
---------+------------------------------ F( 3, 41) = 5.70
Model | 355.782627 3 118.594209 Prob > F = 0.0023
Residual | 853.417373 41 20.8150579 R-squared = 0.2942
---------+------------------------------ Adj R-squared = 0.2426
Total | 1209.20 44 27.4818182 Root MSE = 4.5624
------------------------------------------------------------------------------
conform | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
fscore | -.1510988 .0717105 -2.107 0.041 -.2959211 -.0062766
d | -15.53408 4.400445 -3.530 0.001 -24.42096 -6.647198
intfd | .2611023 .0969992 2.692 0.010 .0652084 .4569961
_cons | 20.79348 3.262732 6.373 0.000 14.20425 27.3827
------------------------------------------------------------------------------
Result on page 194 using the same data file as above.
gen s=1 if(status==2)
(23 missing values generated)
replace s=-1 if(status==1)
(23 real changes made)
gen intfs=fscore*s
reg conform fscore s intfs
Source | SS df MS Number of obs = 45
---------+------------------------------ F( 3, 41) = 5.70
Model | 355.782627 3 118.594209 Prob > F = 0.0023
Residual | 853.417373 41 20.8150579 R-squared = 0.2942
---------+------------------------------ Adj R-squared = 0.2426
Total | 1209.20 44 27.4818182 Root MSE = 4.5624
------------------------------------------------------------------------------
conform | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
fscore | -.0205477 .0484996 -0.424 0.674 -.1184946 .0773992
s | -7.767039 2.200223 -3.530 0.001 -12.21048 -3.323599
intfs | .1305511 .0484996 2.692 0.010 .0326042 .2284981
_cons | 13.02644 2.200223 5.921 0.000 8.582997 17.46988
------------------------------------------------------------------------------
Table in the middle of page 197 using teh data file friendly.
use http://www.ats.ucla.edu/stat/stata/examples/ara/friendly, clear (From Fox, Applied Regression Analysis. Use 'notes' command for source of data ) sort cond by cond: summarize correct -> cond=Before Variable | Obs Mean Std. Dev. Min Max ---------+----------------------------------------------------- correct | 10 36.6 5.337498 24 40 -> cond=Meshed Variable | Obs Mean Std. Dev. Min Max ---------+----------------------------------------------------- correct | 10 36.6 3.025815 30 40 -> cond=SFR Variable | Obs Mean Std. Dev. Min Max ---------+----------------------------------------------------- correct | 10 30.3 7.334091 21 39
Figure 8.8 on page 198.
use http://www.ats.ucla.edu/stat/stata/examples/ara/friendly, clear (From Fox, Applied Regression Analysis. Use 'notes' command for source of data ) egen cm=mean(correct), by(cond) encode cond, gen (x) graph twoway (scatter correct x, jitter(5)) (scatter cm x, connect(l) sort), xlabel(1 2 3)
Table at bottom of page 199 First we do the encoding based on the scheme on page 198.
use http://www.ats.ucla.edu/stat/stata/examples/ara/friendly, clear
(From Fox, Applied Regression Analysis. Use 'notes' command for source of data )
gen c1=1 if(cond=="SFR")
(20 missing values generated)
gen c2=0 if(cond=="SFR")
(20 missing values generated)
replace c1=-1/2 if(cond=="Before")
(10 real changes made)
replace c2=1 if(cond=="Before")
(10 real changes made)
replace c1=-1/2 if(cond=="Meshed")
(10 real changes made)
replace c2=-1 if(cond=="Meshed")
(10 real changes made)
reg correct c1 c2
Source | SS df MS Number of obs = 30
---------+------------------------------ F( 2, 27) = 4.34
Model | 264.60 2 132.30 Prob > F = 0.0232
Residual | 822.90 27 30.4777778 R-squared = 0.2433
---------+------------------------------ Adj R-squared = 0.1873
Total | 1087.50 29 37.50 Root MSE = 5.5207
------------------------------------------------------------------------------
correct | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
c1 | -4.2 1.42543 -2.946 0.007 -7.124742 -1.275258
c2 | 0 1.234459 0.000 1.000 -2.532901 2.532901
_cons | 34.5 1.007932 34.229 0.000 32.4319 36.5681
------------------------------------------------------------------------------
anova correct c1 c2, se cont(c1 c2)
Number of obs = 30 R-squared = 0.2433
Root MSE = 5.52067 Adj R-squared = 0.1873
Source | Seq. SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 264.60 2 132.30 4.34 0.0232
|
c1 | 264.60 1 264.60 8.68 0.0065
c2 | 0.00 1 0.00 0.00 1.0000
|
Residual | 822.90 27 30.4777778
-----------+----------------------------------------------------
Total | 1087.50 29 37.50
The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California.