mvpa - multivariate pattern analysis
Datasets
Download options
Inspection
Principal component analysis
Covariate projection
PLS-R / Selectivty Ratios
📖 Online documentation
Load dataset
Browse...
Select method
Transpose dataset
Select a subset of the current dataset
Missing value methods
Remove invariant variables
String value methods
Introduce offset to avoid negative or zero values
Introduce unique variable combinations
Scaling / Normalization methods
Apply
Select variable for subsetting
Remove or impute missing values
Impute
Remove variables
Remove objects
Missing value impute method
Variable mean
Variable median
Custom value
Enter imputation value
Apply
Apply
Apply
Indicate which level means '1'
Apply
Current invariant variables
Apply
Set new minimum value
Apply
Select function
log10
log2
ln
nth root
Standardization
Normalize to max
Min-max scaling
Choose n (default 2 -> square root)
Ignore response variable
Apply
Select method:
Multiplication ( * )
Addition ( + )
Substraction ( - )
Division ( / )
Ignore response
Apply
Name of new dataset
Select response
Verify and store current dataset
Available valid datasets
Use as current dataset
Consider dataset deletion
Confirm deletion
Load demo dataset
Current dataset
Valid dataset
Download table
Remove
Keep
Undo last operation
Reset dataset
Download table
Show statistics
Select table download format:
tab-delimited values (.tsv)
comma-delimited values (.csv)
Excel table (.xlsx)
Select valid dataset
Change font size
Normality check / QQ-plot
Variable correlation
Select valid dataset
Ignore variables
Standardize
Perform principal component analysis
Change font size
Scree plot
Scores plot
Loadings plot
Variable variation plot
Select valid dataset
Select covariate(s)
Standardize
Perform covariate projection
Dataset after covariate projection
Explained variance by covariate - Plot
Download table
Select valid dataset
Select reponse
Select number of components
Perform PLS-R
Monte Carlo resampling
Standardize
Monte Carlo resampling options
Use seed for reproducibility
Standard deviation based on full dataset
Number of repetitions
Proportion of objects in calibration dataset (%)
Validation threshold
Cost function
Root-mean-square error of prediction (RMSEP)
Mean absolute error (MAE)
Model information
Target projection - Repeated sampling
Target projection - Full dataset
Target projection - Variable variance distribution