search

JMSLab / xtevent

Stata package -xtevent-
MIT License
44 stars 12 forks source link

Wording of warning #116

Closed jmshapir closed 2 years ago

jmshapir commented 2 years ago

When anticipatory effects are allowed we display a warning indicating which coefficient we normalize:

https://github.com/JMSLab/xtevent/blob/30656631e4dfe55c9aa8e0881184bd290167c9d3/xtevent/xtevent.ado#L174

I think the language "normalized to N" for N a lead may be confusing since it may suggest we're normalizing the coefficient to be equal to the value N rather than that we're normalizing (to a value of zero) the coefficient corresponding to lead N.

Maybe we should replace

so the coefficients were normalized to `norm'.

with something like

... so the coefficient at norm was selected to be normalized to zero

to be more in line with language elsewhere e.g.

https://github.com/JMSLab/xtevent/blob/30656631e4dfe55c9aa8e0881184bd290167c9d3/xtevent/xtevent.ado#L180

?

jorpppp commented 2 years ago

Sounds good, thanks @jmshapir. @Constantino-Carreto-Romero , can you implement please?

Constantino-Carreto-Romero commented 2 years ago

In https://github.com/JMSLab/xtevent/commit/6c9327a1471126b36ad806d56e026d3224f528e4 I edited the message indicated in https://github.com/JMSLab/xtevent/issues/116#issue-1413793666. I checked in other parts of the ado files for other messages that could need to be changed as well, but I didn't find such messages. The new message in #114 considers this wording.