Data structures

Data structures store the information on which a GenStat program operates. Structures can be defined, or declared, by a GenStat statement known as a declaration. The directive for declaring each type of structure has the same name as given to that type of structure, for example SCALAR to declare a scalar (or single-valued numerical structure), and so on. These are the directives, with details of their corresponding data structures:

single number

series of numbers

series of character strings (or lines of text)

series of group allocations (using a pre-defined set of numbers or strings to indicate the groups)

rectangular matrix

SYMMETRICMATRIX

symmetric matrix

diagonal matrix

table (to store tabular summaries like means, totals etc)

single identifier

series of identifiers (e.g. to represent a set of structures)

arithmetic expression

model formula (to be fitted in a statistical analysis)

latent roots and vectors

sums of squares and products with associated information such as means

tree (as used to represent classification trees, identification keys and regression trees)

model for Box-Jenkins modelling of time series

It is possible to declare new structures with attributes the same as those of an existing structure.

forms new data structures with attributes taken from an existing structure

duplicates a pointer, with all its components

You can also define data structures whose contents are customized for particular tasks.

defines a customized data structure

declares one or more customized data structures