Transformations

Introduction

Transformation is the process of modifying the data (for example, dividing all of the data values in a field by 10) or deriving new data for analysis (for example, summing the data in two fields). Sometimes it will be easier to use a spreadsheet to calculate derived variables than to modify variables using ASReml transformations.

Transformation qualifiers are listed after data field labels (and the fieldtype if present). They define an operation (e.g. +), often involving an argument (a constant or another variable), which is performed on a target variable. The target is usually implicit, the current field, but can be changed to a new variable with the !TARGET qualifier.

Using transformations will be easier if you understand the process. As ASReml parses the variable definitions, it sequentially assigns them column positions in the internal data array. \Rev{08}It notes which is the last variable which is not created by transformation, and that determines how many fields are read from the data file (unless overridden by the !READ qualifier argument). After parsing the model line, ASReml actually reads the data file. It reads a line into a temporary vector, performs the transformations in that vector, and then saves the positions that relate to labelled variables to the internal data array.

Note that in ASReml
  • alpha and integer fields are converted to numbers before transformations are applied,
  • transformations may be applied to any data field since all data fields are stored as real numbers, but it may not be sensible to change factor level codes,
  • all fields in a record are read before transformations are performed,
  • transformations are performed in the order of appearance.

    • Syntax

    Transformations have one of six forms, namely
    Forms Function Example
    !o to perform an operation on the current field !ABS
    to take absolute values,
    !o v to perform an operation involving an argument on the current field, !+3
    to add 3 to all elements in the field,
    !oVn to perform an operation o on the current field using the data in another field ( v ) !-V2 to subtract field 2 from the current field.
    !TARGET t to change the field modified by subsequent transformations, !TARGET 12
    changes focus to variable 12
    !Vt=c to change all of the data in a target field to a given value, !V12=1.
    !Vt=Vn copy field n to field t; special case is !Vt=V0 instructing ASReml to put the record number into the target !V12=V3
    In these definitions, V is the literal captital V character. The combination Vn can be replaced by the LABEL of the field if the field is already labelled.

    Rules

  • ! flags the presence of the transformation
  • o is the operator or function name,
  • v is the argument required by the transformation,
  • V is a literal character and is followed by the number t (target) or n (argument) of the data field; the data field is used or modified depending on the context,
  • Vt or Vn may be replaced by the name of the field if the field is already named.
  • in the first three forms the operation is performed on
    • the current field; this will be the field associated with the label unless this field has been reset by specifying a new target in a preceding transformation,
  • the fourth form changes the focus
    • for subsequent transformations to the field target,
  • in the last two forms the operation
    • is performed on the target field which is modified as a consequence, for example, to create a copy of field 11, which is an existing data field, we would use 
     !V22=V11
    which copies field 11 into field 22; if field 22 does not exist in the data it is created; note that subsequent transformations on the line will also apply to the target ,
  • ASReml expects to read values for each variable from the data file unless the variable
    • and all subsequent ones are created using != . The number of variables read can be explicitly set using the !READ qualifier

    Transformation functions

  • arithmetic operators = + - * / ^ == <= >= <>
  • Algebraic operators ABS ARCSIN SIN COS EXP MIN MAX MOD
  • genetic marker operators MMAP DOM
  • date operators Jddm Jmmd Jyyd Jddy
  • Drop records D
  • missing value operators M NA
  • substitution operators SEQ SET SUB REPLACE RESCALE SPLIT
  • distribution operators SEED NORMAL UNIFORM
  • repeating operations DO ENDDO

    Return to start