1. Home
  2. Estimate Missing Values

Estimate Missing Values

Use this dialog to estimate missing values in columns, so that seasonal totals or means can be formed. You need to specify the columns whose missing values are to be estimated, and the analysis model to fit their missing values. After the missing values have been replaced by their estimates, their cells will have bookmarks assigned to them in case you want to revert them back to missing.

Columns to estimate values for

This selects the columns whose missing values are to be estimated.

Current column only the column that had focus when opening this menu.
All trial columns all the trial columns in the spreadsheet with missing values.
Specified column types only columns in the spreadsheet of the type selected in the Estimate in columns of type drop-down list.

Blocking model in missing value analysis

Automatic analysis use the VAROWCOLUMNDESIGN procedure to find the optimal spatial model.
Replicates (randomized block) use analysis of variance with only replicates as the block term.
Replicates + columns use REML analysis with replicates and columns as the spatial terms.
Replicates/columns use REML analysis with replicates and columns within replicates as the spatial terms.
Rows + columns use REML analysis with rows and columns as the spatial terms.
Nearest neighbour (1d – columns) use REML with an autocorrelation of order 1 along columns as the spatial term.
Nearest neighbour 2d (rows x cols) use REML analysis with an autocorrelation of order 1 along rows and columns as the spatial terms.

The default method of Automatic analysis is recommended as this will revert to a simpler model if full spatial detrending is not required.

Action buttons

Ok Run the missing value estimation.
Cancel Close the dialog without further changes.

See also

Updated on April 19, 2024

Was this article helpful?