NEWS.md

In metinbulus/cosa: Bound Constrained Optimal Sample Size Allocation

Changes in the cosa 2.1.0

• Functions can accomodate polynomial functional forms for the score variable w/ or w/o interaction with the treatment variable
• Improvements to bound constrained optimization (e.g. removed two-stage optimization)

Changes in the cosa 2.0.0

• Functions can accomodate quadratic functional forms for the score variable
• Users can override default degrees of freedom
• A few minor changes to naming conventions

Changes in the cosa 1.2.2

• Added vectorize.cosa() function, which makes it convenient to run and compare multiple designs
• Removed MMA and COBYLA optimizers from default options

Changes in the cosa 1.2.1

• Removed experimental wrapper functions cosa(), mdes() and power()
• Summaries are printed with the function call
• Users can specify benchmark values in plots
• Changed the package title

Changes in the cosa 1.2.0

• Added jacobians
• Option for maximizing power when primary constraint is placed on the total cost
• Bug fixes for cosa.bird2r1() function

Changes in the cosa 1.1.0

• Added experimental wrapper functions cosa(), mdes() and power()
• Added designs with top-level stratification (fixed blocks)
• Bug fixes for bcrd4r2 and bcrd4r3 designs
• Added S3 plot() methods
• Improved validity checks for arguments
• Added citation

metinbulus/cosa documentation built on Sept. 9, 2021, 12:04 p.m.