|
|
|
||||
|
|
|||||
Table 10.1, page 265.
use http://www.ats.ucla.edu/stat/stata/examples/chp/p233, clear drop if year >= 60 regress import doprod stock consum, beta Source | SS df MS Number of obs = 11 ---------+------------------------------ F( 3, 7) = 285.61 Model | 204.776154 3 68.2587179 Prob > F = 0.0000 Residual | 1.67295319 7 .238993312 R-squared = 0.9919 ---------+------------------------------ Adj R-squared = 0.9884 Total | 206.449107 10 20.6449107 Root MSE = .48887 ------------------------------------------------------------------------------ import | Coef. Std. Err. t P>|t| Beta ---------+-------------------------------------------------------------------- doprod | -.0513959 .0702801 -0.731 0.488 -.3393412 stock | .5869492 .0946185 6.203 0.000 .2130485 consum | .2868483 .1022083 2.807 0.026 1.30268 _cons | -10.12798 1.212161 -8.355 0.000 . ------------------------------------------------------------------------------
Equations 10.6 and 10.7, page 265.
Note: The signs of the coefficients for the third eigenvector are different from those given in 10.7. This difference is merely a reflection of the signs and does not change the meaning of the analysis.
factor doprod stock consum, pc factors(3)
(obs=11)
(principal components; 3 components retained)
Component Eigenvalue Difference Proportion Cumulative
------------------------------------------------------------------
1 1.99915 1.00100 0.6664 0.6664
2 0.99815 0.99546 0.3327 0.9991
3 0.00269 . 0.0009 1.0000
Eigenvectors
Variable | 1 2 3
----------+--------------------------------
doprod | 0.70633 -0.03569 0.70698
stock | 0.04350 0.99903 0.00697
consum | 0.70654 -0.02583 -0.70720
score c1 c2 c3
(based on unrotated principal components)
Scoring Coefficients
Variable | 1 2 3
----------+--------------------------------
doprod | 0.70633 -0.03569 0.70698
stock | 0.04350 0.99903 0.00697
consum | 0.70654 -0.02583 -0.70720
matrix ld = get(Ld) /* save loading matrix for later analyses */
Table 10.2, page 265.
Note: The variable import has been transformed into a standard score zimport.
egen zimport = std(import)
regress z c1 c2 c3, noconst
Source | SS df MS Number of obs = 11
---------+------------------------------ F( 3, 8) = 326.41
Model | 9.91896497 3 3.30632166 Prob > F = 0.0000
Residual | .081034631 8 .010129329 R-squared = 0.9919
---------+------------------------------ Adj R-squared = 0.9889
Total | 9.9999996 11 .909090872 Root MSE = .10064
------------------------------------------------------------------------------
zimport | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------+--------------------------------------------------------------------
c1 | .6899821 .0225096 30.653 0.000 .6380749 .7418892
c2 | .1913035 .031856 6.005 0.000 .1178434 .2647636
c3 | -1.159675 .6135396 -1.890 0.095 -2.5745 .2551501
------------------------------------------------------------------------------
Standardized coefficients for Table 10.3, page 268.
Note 1: We will use Stata's matrix commands for these computations.
matrix b = get(_b)
matrix v = b'
matrix v[2,1] = 0
matrix v[3,1] = 0
matrix list v
v[3,1]
y1
c1 .68998207
c2 0
c3 0
matrix pc1 = ld*v
matrix list pc1 /* standardized coefficients for first pc */
pc1[3,1]
y1
r1 .48735532
r2 .03001462
r3 .487503
matrix v = b'
matrix v[3,1] = 0
matrix list v
v[3,1]
y1
c1 .68998207
c2 .19130349
c3 0
matrix pc2 = ld*v
matrix list pc2 /* standardized coefficients for first and second pc */
pc2[3,1]
y1
r1 .48052795
r2 .22113237
r3 .48256155
matrix v = b'
matrix list v
v[3,1]
y1
c1 .68998207
c2 .19130349
c3 -1.1596748
matrix pc3 = ld*v
matrix list pc3 /* standardized coefficients for all pcs */
pc3[3,1]
y1
r1 -.33934136
r2 .21304848
r3 1.3026802
use http://www.ats.ucla.edu/stat/stata/examples/chp/p270, clear
list
u c1 c2 c3 c4
1. .955 1.467 1.903 -.53 .0389
2. -.746 2.136 .238 -.29 -.03
3. -2.323 -1.13 .184 -.01 -.094
4. -.82 .66 1.577 .179 -.033
5. .471 -.359 .484 -.74 .019
6. -.299 -.967 .17 .086 -.012
7. .21 -.931 -2.135 -.173 .008
8. .558 2.232 -.692 .46 .023
9. -1.119 .352 -1.432 -.032 -.045
10. .496 -1.663 1.828 .851 .02
11. .781 1.641 -1.295 .494 .031
12. .918 -1.693 -.392 -.02 .037
13. .918 -1.746 -.438 -.275 .037
Table 10.5, page 270.
regress u c1 c2 c3 c4
Source | SS df MS Number of obs = 13
-------------+------------------------------ F( 4, 8) =63662.46
Model | 11.9971847 4 2.99929618 Prob > F = 0.0000
Residual | .0003769 8 .000047112 R-squared = 1.0000
-------------+------------------------------ Adj R-squared = 1.0000
Total | 11.9975616 12 .999796801 Root MSE = .00686
------------------------------------------------------------------------------
u | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
c1 | -.0017999 .001325 -1.36 0.211 -.0048552 .0012555
c2 | -.0025529 .0015783 -1.62 0.144 -.0061924 .0010866
c3 | .0016481 .0045872 0.36 0.729 -.0089299 .0122262
c4 | 24.7659 .0490776 504.63 0.000 24.65273 24.87908
_cons | .0001904 .0019037 0.10 0.923 -.0041995 .0045803
------------------------------------------------------------------------------
Table 10.6, page 270.
regress u c1 c2 c3
Source | SS df MS Number of obs = 13
-------------+------------------------------ F( 3, 9) = 0.00
Model | .000054633 3 .000018211 Prob > F = 1.0000
Residual | 11.997507 9 1.33305633 R-squared = 0.0000
-------------+------------------------------ Adj R-squared = -0.3333
Total | 11.9975616 12 .999796801 Root MSE = 1.1546
------------------------------------------------------------------------------
u | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
c1 | -.0012226 .2228739 -0.00 0.996 -.5053984 .5029532
c2 | -.0001243 .2654817 -0.00 1.000 -.6006856 .6004371
c3 | .0025214 .7716171 0.00 0.997 -1.742998 1.748041
_cons | -8.83e-08 .320223 -0.00 1.000 -.7243949 .7243948
------------------------------------------------------------------------------
Figure 10.1, page 271.
graph twoway scatter u c1, nodraw ylabel(-2(1)1) xlabel(-1(1)2) saving(f101) graph twoway scatter u c2, nodraw ylabel(-2(1)1) xlabel(-2(1)2) saving(f102) graph twoway scatter u c3, nodraw ylabel(-2(1)1) xlabel(-.4(.4).8) saving(f103) graph twoway scatter u c4, nodraw ylabel(-2(1)1) xlabel(-.08(.04).04) saving(f104) graph combine f101.gph f102.gph f103.gph f104.gph
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services
