Checking for NAs in rolling spillover estimation

Some users were complaining about cryptic errors that would typically
boil down to having NAs in the data in rolling estimation.

This commit amends the rolling spillovers to check for the NAs and
fail if it finds any.

See the relevant commit

Fixed documentation errors from CRAN

0.2.2

Fixed documentation errors

Version 0.2.1

Small fixes, mainly references updated to point to the published methodology.

Version 0.2.0

Complete rewrite of the package, new and better functionality, see the readme.

Bug release

The BK did not sum to the original DY measure due to a typo. It has been fixed and now there is a test case checking for summation.

Version released with working paper.

This release was used to compute the results in the WP Barunik, Krehlik (2015). Future releases should contain potential minor fixes of potential bugs.