Transformations update the DataSet with some kind of mathematic function. These transformations can be accessed from the table menu. There are different transformations available for observations and variables. The transformations are available directly from the table menu and they can also be selected from the New Modification Wizard.

Right-clicking on a transformation in the DataSet table will bring up the table menu where the first option is a menu for that particular transformation. The name of the transformation will be shown in italic.

Transformation menu options:

Include: Includes the selected observations/variables for the particular transformation.

Exclude: Exclude the selected observations/variables for the particular transformation.

Delete modification: Removes the transformation from the DataSet.

Rename modification: Renames the transformation.

Variable transformations

- Center: Performs variable mean centering, which means the variable means are deducted from the values.

- UV scale: Performs Unit-Variance scaling, which means all variable have equal variance of one unit. Each value is divided by the standard deviation of the actual variable.

- Pareto: Performs Pareto-scaling, which means all values are divided by the square-root of the standard deviations of the variables.

- Quad term: Adds quadratic terms for the selected variables. The new transformed variables will appear at the bottom of the DataSet table.

- Interaction term: Adds interaction terms for the selected variables. Only two-way interaction terms are available in Evince. The new transformed variables will appear at the bottom of the DataSet table.

- Function: Adds a custom function from the selected variables. The user must enter the function in the Settings Panel to the right of the Evince workspace. The selected variables are coded to alphabetic letters, which are available for the user to use in the custom expression. The letters available will be shown in the Settings Panel below the expression. The new transformed variable will appear at the bottom of the DataSet table.

- Logarithm: Adds the logarithms of the selected variables. The user can enter a constant in the settings panel in order to prevent that logarithms are performed for values of zero or less.

- MSC Transform: Performs Multiplicative Scatter Correction. This transformation is useful for spectral observations that exhibit large baseline shifts and large variations in slope. The MSC regresses each spectrum against the average spectrum of the training set:

Each individual spectrum can then be corrected with the constants from the above equation:

- Derivative: Calculates the first-derivative for all observations. Each value is calculated as:

- Savitzky-Golay: Performs Savitzky-Golay smoothing and differentiation on all observations. This implies that the signal for each observation is smoothed according to a window size that the user can specify in the Settings Panel to the right of the Evince workspace. The user must give the polynomial order of the equation that is to be used for the smoothing. A higher polynomial order will ensure a better fit. The user also has to give the wanted derivative order. Please note that the derivative order cannot be higher than the polynomial order. Finally, the user has to give the number of left and right points, which is equivalent to the window size used for the smoothing.

- SNV Correction: Similar to the MSC transform, the Standard Normal Variate, SNV, transformation corrects all observations with respect to baseline shifts and slopes. SNV implies that all observations are centered and scaled:

Transformations applicable to both observations and variables

Reduce to Average: The selected observations/variables are reduced to an average. The user must specify the group size in the Settings Panel. The group size is equivalent to the number of observations/variables that are averaged into a single mean. The mean observations/variables can be found at the bottom of the DataSet table.