You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We sometimes use horizon interchangeably with ahead and so perhaps we should change the name of it in step_growth_rate() and arx_class_args_list() (and anywhere else it gets used in such a way in this package). Since it is passed to the h argument of epiprocess::growth_rate(), then maybe the easiest thing is to just refer to it as h?
The text was updated successfully, but these errors were encountered:
Since horizon is used by the Hubverse for "reference date"-relative targeting (which are something like a label for (the end of) some time interval containing the forecast submission deadline), perhaps we want to stay away from that term and h for these, and standardize around ahead, at least for max time_value-relative stuff?
(The reasoning behind these reference dates includes allowing some flexibility for forecast deadlines to change without changing horizon-labeling and without having people have to be aware of when to pretend forecasts were submitted on a different date. E.g., the time intervals used were epiweeks, so changing deadline from say Tue -> Wed would not impact the reference date.)
We sometimes use horizon interchangeably with ahead and so perhaps we should change the name of it in
step_growth_rate()
andarx_class_args_list()
(and anywhere else it gets used in such a way in this package). Since it is passed to theh
argument ofepiprocess::growth_rate()
, then maybe the easiest thing is to just refer to it ash
?The text was updated successfully, but these errors were encountered: