Improved treatment of doubling / halving time when growth rate estimate spans 0#94

Open

jrcpulliam wants to merge 12 commits into

reconhub:mainfrom

jrcpulliam:master

Commits on Dec 12, 2018

add skeleton of test for #28
jrcpulliam
committed

Commits on Dec 13, 2018

add test to ensure that CI of doubling / halving times do not have negative values
jrcpulliam
committed
passes my tests for #28, but breaks other tests (on first run of tests()) and regenerates reference files; passes all tests on second run of tests(); not committing new reference files
jrcpulliam
committed
change all occurrences of expect_equal_to_reference() to expect_known_value() with update=FALSE; note that there is a bug in expect_equal_to_reference() such that update is reset from FALSE to TRUE…
jrcpulliam
committed
updating test for handling groups for r estimates with different signs; need input from @thibautjombart on what expected behavior should be
jrcpulliam
committed
have reviewed errors and confirmed that the new values reflect the desired behaviour as a result of the change to doubling / halving interval CIs; committing new reference rds files
jrcpulliam
committed
throw error if growth rates for groups have different signs but are fit together; add test for this
jrcpulliam
committed
adjust data generation for some automated tests so not relying on groups that have growth rates with different signs
jrcpulliam
committed
regenerate reference fits due to change in data used to generate the fits (see previous commit)
jrcpulliam
committed
fixing spaces to make tibaut happy
jrcpulliam
committed
update NEWS.md to reflect changes; closes #28
jrcpulliam
committed
minor; mostly adding more spaces to make tibaut happier
jrcpulliam
committed