>
|
|
The ScreePlot command returns a plot of the proportion of the total variance which is attributable to each of the components in a principal component analysis.
>
|
|
Values proportion of variance St. Deviation
8.3208 0.8782 2.8846
1.1119 0.1173 1.0545
0.0425 0.0045 0.2061
| |
The Scree Plot plots the values taken from PCAnalysis:-values and PCAnalysis:-varianceproportion as seen above.
From the above Scree plot, the amount of variation explained by each component drops significantly after the first component. This may suggest that using one component may be enough to summarize the data.
>
|
|
In the case that a 2 column or larger dataset is given as the argument to ScreePlot, the principal component analysis is computed and the plot is returned.
>
|
|
The above plot shows that the first 3 components account for 99.5% of the variance.
It may not always be possible to summarize a dataset using only a few components.
>
|
|
>
|
|
>
|
|