Algorithms

Solving Linear Mixed model Equations

lmt supports two types of solver for solving MME's: a direct solver and an iterative solver

Iterative solver

The iterative solver uses the preconditioned conjugate gradient method and is lmt's default solver. It does not require the explicit construction of any mixed model equation, and is therefore less resource demanding than the direct solver. That is, many models which cannot be solved using the direct solver can still be solved using the iterative solver. Even for small models the iterative solver usually outperforms the direct solver in terms of total processing time.

Whether the iterative solver has converged in round $$i$$ can be evaluated with convergence criterions $$log_e\left(\sqrt{\frac{||(Cx_i-b)||}{||b||}}\right)<t$$ or $$log_e\left(\sqrt{\frac{||(x_{i}-x_{i-1})||'}{||x_{i-1}||}}\right)<t$$, where $$C$$ is the mixed-model coefficient matrix, $$x_i$$ is the solution vector in round $$i$$, $$b$$ is the right-hand side and $$t$$ is the convergence threshold which defaults to -18.42, which is $$log_e(10^{-9})$$.

Direct solver

The direct solver requires the mixed model coefficient matrix to be build and all Kronecker products to be resolved. This can be quite memory demanding and should therefore be used carefully. The direct solver uses a Cholesky decomposition and forward-backward-substitution to solve the mixed model equation system, where especially the decomposition step can be very resource demanding and time consuming.

Variance component estimation

For random factors lmt supports variance of structure $$\Gamma\otimes\Sigma$$, where $$\Sigma$$ is an dense symmetric positive definite matrix to be estimated. For residuals lmt supports variance structures $$I\otimes\Sigma$$ and $$\Theta_L(I_{n_{observations}})\Theta_L^{'}$$, where $$\Theta$$ is symmetric positive definite block-diagonal matrix of $$n$$ symmetric positive definite martices $$\Sigma_i, i=1,..,n$$, $$\Theta_L$$ is the lower Cholesky factor of $$\Theta$$ and $$I_{n_{observations}}$$ is an identity matrix of dimensions equal to the total number of observations across all traits. Note that the number of records associated to a particular $$\Sigma_i$$ should be sufficient to facilitate its estimation.

Gibbs sampling

Single pass Gibbs sampling

lmt's single pass Gibbs sampling algorithm is described in . In short, all location parameters are drawn from their joint conditional posterior distribution. Note that this requires solving the mixed model equation system once per iteration which usually leads to a substantial increase in processing time. Note that ssSNBPLUP models are not supported.

Blocked Gibbs sampling

For random factors lmt's blocked Gibbs sampler draws correlated location parameters within factor level from their joint conditional posterior distribution. Location parameters of fixed factors are drawn in scalar mode from their fully conditional posterior. Note that ssGTBLUP and ssSNPBLUP models are not supported.

Restricted Maximum Likelyhood

MC-EM-REML

lmt provides a monte-carlo expectation-maximisation REML algorithms which uses the preconditioned gradient solver for solving the mixed model equations and a blocked Gibbs sampler to sample the necessary traces. Note that ssSNPBLUP and ssGTBLUP models are not supported.

Average information (AI)-REML

lmt provides the calculation of variance components using average information REML , and . REML estimates of co-variance matrices can be derived using the phenotypic co-variance matrix $$V$$ or the mixed-model equations system coefficient matrix C.

lmt provides three different AI-REML convergence criterions:

• the relative change of the log-likelihood calculated as $$log_e\left(\sqrt{\frac{||(l_{i}-l_{i-1})||'}{||l_{i-1}||}}\right)$$ where $$l$$ is the log-likelihood and $$i$$ is the iteration counter.
• $$log_e\left(\sqrt{\frac{||(p_{i}-p_{i-1})||'}{||p_{i-1}||}}\right)$$ where $$p$$ is the parameter vector and $$i$$ is the iteration counter.
• $$log_e\left(\sqrt{||g_{i}||}\right)$$ where $$g$$ is the gradient vector and $$i$$ is the iteration counter.

AI-REML-C

lmt supports AI-REML-C, which relies on the construction and factorization of the mixed-model equations system coefficient matrix C.

Note that ssSNPBLUP and ssGTBLUP models are not supported. Further, it is not advisable to use airemlc for ssGBLUP models.

Fixation of $$\Sigma$$ matrix elements

Elements of $$\Sigma$$ matrices can be exempted from re-estimation in two ways:

1. providing a boolean mask matrix $$B$$ where elements set to "T" are related to elements in $$\Sigma$$ which should be regarded as fixed, or by
2. setting a diagonal element in $$\Sigma$$ desired to be fixed to 1.0 to 1.0, or by
3. setting an off-diagonal element in $$\Sigma$$ desired to be fixed to 0.0 to 0.0.

Note that in case exemption is communicated via options 2 and 3 the $$\Sigma$$ matrix provided at start must still be positive definite. Further note that using option 1 overrides all information contained in $$\Sigma$$. That is if $$\Sigma[1,1]$$ is set to 1.0 but $$B$$[1,1] is set to false, $$\Sigma$$[1,1] is not exempt.

Elements of the inverse of the mixed model coefficient matrix

In principle lmt can generate any element of the inverse mixed model coefficient matrix. However, the user interface is currently limited to the diagonal elements for fixed factors and the diagonal blocks for random factors. These elements can either be sampled or obtained accurately via solving.

Gibbs Sampling

Following the approach of Harville(1999) lmt can sample for fixed factors the diagonal elements of the inverse of the mixed model coefficient matrix, for random factors the diagonal blocks of the inverse of the coefficient matrix where the block size is determined by the dimension of the related $$\Sigma$$ matrix. The blocks are the prediction error co-variance matrices of the factor levels of correlated sub-factors. When sampling prediction error variances lmt can run many Gibbs chains in parallel allowing to exploit multi-core hardware architecture. However, it is recommended to specify not more chains than the number of available real cores excluding hyper-threading technology.

Solving

lmt can obtain elements of the inverse of the coefficient matrix via solving the mixed model equations. This method is currently only supported for the diagonal prediction error co-variance blocks of random factors, where the block size is determined by the dimension of the related $$\Sigma$$ matrix. For this algorithm lmt can utilize either the #Iterative solver or the #Direct solver.

Iterative inbreeding

lmt supports the iterative calculation of inbreeding coefficients as described in VanRaden(1992).