deconLNR {SpatialDecon} | R Documentation |
Calls lognlm() to optimize the model.
deconLNR(Y, X, bg = 0, weights = NULL, epsilon = NULL, maxit = 1000)
Y |
p-length expression vector or p * N expression matrix - the actual (linear-scale) data |
X |
p * K Training matrix |
bg |
scalar or matrix of expected background counts per data point. |
weights |
The same as the weights argument used by lm |
epsilon |
optional, a very small non-zero number to use as a lower threshold to make fits well-behaved |
maxit |
Maximum number of iterations. Default 1000. |
a list: beta (estimate), sigmas (covariance matrix of estimate, derived by inverting the hessian from lognlm)