Table 1 Prior densities available for regression coefficients in the BGLR package
Model (prior density)HyperparametersTreatment in BGLRa
Flat (FIXED)Mean (μβ)μβ = 0
Variance Embedded ImageEmbedded Image
Gaussian (BRR)Mean (μβ)μβ = 0
Variance Embedded ImageEmbedded Image
Scaled-t (BayesA)Degrees of freedom (dfβ)User specified (default value, 5)
Scale (Sβ)SβGamma
Double exponential (BL)λ2λ fixed, user specified, or 
λ2Gamma, or 
Embedded Imageb
Gaussian mixture (BayesC)π (prop. of nonnull effects)πBeta
dfβUser specified (default value, 5)
SβSβGamma
Scaled-t mixture (BayesB)π (prop. of nonnull effects)πBeta
dfβUser specified (default value, 5)
SβSβGamma
  • a Further details are given in the supporting information (Section A of File S1).

  • b This approach is further discussed in de los Campos et al. (2009b).