The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

logbin 2.0.6

• Replace vcov calculation in summary.logbin to replicate the version in summary.glm.
• Implement tests using testthat.
• Clarify the copyright status of functions based on base R.
• (Minor) Removed unused contr.isototonic.rev function. Also no longer exporting interpret.logbin.smooth.
• (Minor) Suppress startup messages when turboEM::turboem is called for the first time.
• (Minor) Update logbin-package Rd file to remove CRAN NOTE.

logbin 2.0.5

• Fixed reparameterisation for factor variables in method = "cem" (@jakobschoepe, #3)
• Implemented a more efficient approach to parameter expansion (using the transformation matrix).
• When using the parameter expansion approach, check convergence of the reduced parameter vector. As an indirect result, the conv.test function is no longer exported.
• Fixed error in logbin if offset provided
• Implement coeftrace via turboEM::turboem argument keep.paramval, to avoid taking a different path when it is turned on (#5)

logbin 2.0.4

• Added citation to Journal of Statistical Software article.

logbin 2.0.3

• Added coeftrace argument to logbin.control, such that coefficient history can be obtained for method = "em" and method = "cem".

logbin 2.0.2

• Changed logbin.ab so that outer.iterations can be set by control$maxit in the call to logbin, and maxit (inner iterations) is set by control.method$maxit.
• logbin now returns contrasts, qr, R and effects components, so associated glm S3 methods (e.g. influence, plot) will work.
• weights component returned is the IRLS working weights at the last iteration, for consistency with glm (even though these are not actually used except with method = "glm").
• plot.logbin.smooth with type = "diagnostics" will plot the model diagnostics usually produced by plot.lm.
• plot.logbin.smooth will allow extra parameters (e.g. col, ylim) to be passed.

logbin 2.0.1

• (Had to increment version number to satisfy CRAN check after correcting an error)

logbin 2.0

• method option added: fit using glm, glm2, adaptive barrier, CEM or single EM algorithm with overparameterised model
• accelerate option added: use turboEM methods to speed up convergence
• (Minor) Corrected the definition of boundary

logbin 1.2

• Minor documentation changes (e.g. updated references)

logbin 1.1

• Very minor documentation changes

logbin 1.0

• Added na.action argument to logbin() and smooth.logbin()

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.