Truncated uniform Kernel function

[JoaoPereira]

Hello

In page 785 in User's guide II Eviews 7 in the table with Kernel function properties about Truncated uniform(1st line) we have a q=0 that don't correspond to the value Rb=1/5 (4th column),for according to the formulae Rb=1/(2q+1) (see below table), q should be 2 and not 0.

To put things clear I estimated a model and calculated the automatic bandwidth that confirms the value in Eviews for q=2.

So is there a bug or am I misunderstanding something?

Thks for assistance

Joao Pereira

[EViews Glenn]

In the second sentence we write "The Truncated kernel does not have theoretically proscribed values...but Andrews (1991) reports Monte Carlo simulations that suggest that these values work well."