  |
|
|
     
|
CycDesigN generates optimal or near-optimal experimental designs with one or two dimensional blocking structures. Both resolvable and non-resolvable block (one-dimensional) and row-column (two-dimensional) designs can be constructed. Single, nested or factorial treatment structures can be specified. In addition, latinized, t-latinized and partially-latinized designs are available for resolvable design with contiguous replicates. There is also a feature that allows the spatial structure of the experiment to be specified, using one of three alternative correlation structures. Apart from the wide range of design types on offer in CycDesigN, there are many other options and facilities such as randomization procedures appropriate to each design type, a log file containing details of the CycDesigN session, a file for the design layout, extensive help facilities, and a comprehensive on-line manual. CycDesigN provides the most comprehensive design generation package yet available for experimenters; particularly those involved in field, glasshouse and laboratory trials. The design types covered incorporate the most recent developments in research into the construction of experimental designs. |
|
Questions or problems regarding this web site
should be directed to David
Whitaker. |
