ASReml

What is ASReml?

Adjust mixed linear models using advanced techniques of maximum restricted likelihood (reml).

Harness the power of REML

ASReml is a powerful statistical software specially designed for mixed models using maximum residual likelihood (REML) to estimate parameters. Mixed-effect linear models provide a rich and flexible tool for analyzing many datasets commonly originated in animal husbandry, plant and aquaculture, agriculture, environmental sciences, and medical sciences.

Using the mean information algorithm (AI) and sparse matrix methods, ASReml handles large or extremely large and complex data analysis (of 500,000 or more observations/effects). It provides flexible methods for modeling a wide range of variance models for random effects or error structures.

Change the text below to read - Typical ASReml applications include analysis of:

• balanced longitudinal data (des)
• balanced (de) designed experiments
• multi-environment testing
• univariate and multivariate animal husbanding
• genetic data
• regular or irregular spatial data
• analysis of repeated measures

Introduction of an alternative functional method of association of variance structures with the terms of the random and residual models, similar to that used in asreml-r, as an alternative to the previous structural method, where variance models were specified separately from the model terms. Using the functional specification, the variance model for the terms of the random model and the residual error term is specified in the mixed linear model involving the terms with the required variance model function. The functional approach leads to a simpler, more concise, and less error-prone specification of the mixed linear model, which is more automatic for specifying residual variances of multiple sections.

There are also many changes to improve efficiency and convenience. These include:

• computationally efficient adjustment of random regression models when there are more variables than observations motivated by the use of SNP marker data to explain genotypes
• adjustment of linear relationships between the parameters of the variance structure
• automatic generation of initial values for variance parameters
• generate a model to allow an alternative way of presenting
• parametric information associated with variance structures
• new qualifiers! ASSIGN! FOR E! IF to simplify workflow
• stabilized updates to improve the convergence of analytical factor models
• improved syntax for VPREDICT, allowing the specification of functions in terms of names instead of numbers
• calculating information criteria
• write design matrices in external files

System requirements:

CPU: 1.8 GHz dual core processor

RAM: 2 GB

HDD / SSD: 50 GB of free space on the main drive

Operating system: Microsoft Windows 8 32-bit, Microsoft Windows 7 32-bit and Microsoft Windows Vista 32-bit

Resolution: 1024×768

Supported operating systems

Mac OS X 10.5 x (Leopard), Mac OS X 10.6 (Snow Leopard), Mac OS X 10.7 (Lion), Mac OS X 10.7 (Mountain Lion), OS X 10.9 (Mavericks) and Mac OS X 10.10 (Yosemite)

Hardware requirements

Memory - 1 GB of RAM and more

Free disk space - 50 MB

System type - 32-bit and 64-bit OS