Вы находитесь на странице: 1из 1

Set more of

cd " C:\User\data_folder
use pca_data.dta, clear
pca var_1 var_2 var_3 var_4 var_5 var_6 var_7 var_8 var_9 var_10 var_11 var_12 var_13 var_14 var_15 var_16 var_17 var_18 var_19 var_20,
mineigen(1.0)
predict pca, score
estat kmo
estat anti
estat loadings
estat summarize
greigen
screeplot, ci
loadingplot
scoreplot
xtile pcaqints=pca, nq(5)
*/
1.
2.
3.

It is important to dichotomise all variables because pca assumes that all variables are in the same units.
Dichotomising therefore achieves this by making the answer category (yes/no as in having the assert or not).
Factors are real factors if their respective eigen value is greater than 1
Therefore, it makes sense to put in the option , mineigen(1.0) this tells stata to calculate all but display components whose minmum eigen
values is 1. 0
Principal components analysis is used for data reduction. The objective of pca, therefore is just to reduce the dimensions of variables that are
measuring the same thing. Thus, pca assumes that several variables were used to measure diferent aspects of the same concept. Since the
variables are measuring the same thing, they must be closely correlated. This is why correlation and covariance are popular in pca. To test
this, the post estimation stata command estat kmo is important to check this assumption. estat kmo displays the Kaiser-Meyer-Olkin (KMO)
measure of sampling adequacy. KMO takes values between 0 and 1, with small values indicating that overall the variables have too little in
common to warrant a PCA analysis. Historically, the following labels are often given to values of KMO (Kaiser 1974):
0.00
0.50
0.60
0.70
0.80
0.90

/*

to
to
to
to
to
to

0.49
0.59
0.69
0.79
0.89
1.00

unacceptable
miserable
mediocre
middling
meritorious
marvelous

Вам также может понравиться