Description Usage Arguments Value Author(s) References See Also Examples
Iterative Principal Components Analysis (IPCA) Computes the Iterative components analysis to find dimensionality of a huge data matrix with entries 0 and 1.
1 2 |
test: |
Dichotomic matrix data. |
h7: |
Threshold to define the items not classificated |
h2: |
Threshold to define the unidimensionality |
hn: |
number of items in each cluster |
angu: |
Threshol to define the angle of each cluster (it could change) |
clas: |
If is TRUE then the classification is made, default is TRUE |
reclas: |
If is TRUE then the reclassification is made, default is TRUE If clas=TRUE and reclas=TRUE then the results are: |
n.submatrix: Total of submatrices Before and After reclassification
n.ipm.befrec: Number of items per matrix before reclassification. If clas=TRUE and reclas=FALSE or clas=FALSE and reclas=FALSE, then this result is NULL.
n.ipm.aftrec: Number of items per matrix after reclassification. If clas=TRUE and reclas=FALSE or clas=FALSE and reclas=FALSE, then this result is NULL.
n.ipm.clas: Number of items per matrix after classification. If clas=FALSE and reclas=TRUE or clas=FALSE and reclas=FALSE, then this result is NULL.
p.axes: Principal Axis of each submatrix
ipm: Matrix of items on each submatrix before classification. If clas=FALSE and reclas=TRUE or clas=FALSE and reclas=FALSE, then this result only show the first classification.
ipmc: Matrix of items on each submatrix after classification. If clas=FALSE and reclas=TRUE or clas=FALSE and reclas=FALSE, then this result is NULL.
DPM: Data of each submatrix (List object)
inc: Items not classified in a principle
SICS Research Group, Universidad Nacional de Colombia ammontenegrod@unal.edu.co
Montenegro, A.M. & Ordo\~nez, M.F & P\'aez, S (2015) A Novel Methodology Based on Item Responde Theory to Find Dimensionality and Emergent Information from Textual Data. Journal of Infometrix(Submited)
1 | irtpp()
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.