Difference between revisions of "Principal Component Analysis (PCA)"
| Line 18: | Line 18: | ||
Thus, the PCA provides us with the information that similar variables have the most in common, but drops the rest. Therefore, the PCA ''really must be used only on variables that have a lot in common, or else a lot of information will be lost''. | Thus, the PCA provides us with the information that similar variables have the most in common, but drops the rest. Therefore, the PCA ''really must be used only on variables that have a lot in common, or else a lot of information will be lost''. | ||
=== | ===How to do it=== | ||
First, check the multiple correlation between your similar variables. Keep only the ones with the highest correlations. Most of the art happens here: The most similar the variables are, the most charged with information the pca is. | |||
Beware of variables with a lot of missings, because the new pca variable will be set as missing for those observations. | |||
Then, the manipulation its simple: you can use the functions pca or pcamat, and predict in Stata. You will want to take a close look at the proportion of the variance that is explained by your first component. You can also use estat kmo (Kaiser-Meyer-Olkin), that tests if your variables were appropriate for factor analysis. | |||
===Subsection 3=== | ===Subsection 3=== | ||
Revision as of 23:32, 7 February 2017
NOTE: this article is only a template. Please add content!
add introductory 1-2 sentences here
Read First
PCA, is a way to create an index from a group of variables that are similar in the information that they provide. This allows maximizing the information we keep, without using variables that will cause multicolinearity, and without having to choose one variables among many.
Guidelines
- organize information on the topic into subsections. for each subsection, include a brief description / overview, with links to articles that provide details
The spatial principle of a PCA
In a space of 3 dimensions, that are for instance income in x, savings in y and consumption in z, we have let’s say 12 vectors that represent our 12 similar variables that were measured in the field. Those vectors combined together create a cloud in 3D. That cloud has 3 principal directions; the first 2 like the sticks of a kite, and a 3rd stick at 90 degrees from the first 2. Well, the longest of the sticks that represent the cloud, is the Principal Component.
Thus, the PCA provides us with the information that similar variables have the most in common, but drops the rest. Therefore, the PCA really must be used only on variables that have a lot in common, or else a lot of information will be lost.
How to do it
First, check the multiple correlation between your similar variables. Keep only the ones with the highest correlations. Most of the art happens here: The most similar the variables are, the most charged with information the pca is.
Beware of variables with a lot of missings, because the new pca variable will be set as missing for those observations.
Then, the manipulation its simple: you can use the functions pca or pcamat, and predict in Stata. You will want to take a close look at the proportion of the variance that is explained by your first component. You can also use estat kmo (Kaiser-Meyer-Olkin), that tests if your variables were appropriate for factor analysis.
Subsection 3
Back to Parent
This article is part of the topic Data Analysis
Additional Resources
- list here other articles related to this topic, with a brief description and link