Fixing minor typo in README

README.md

@@ -167,7 +167,7 @@ All notable changes to this project will be documented below. The format is base
 
 ## Fixed
 - Fixed massive erroneous non-articulation issue in 2025. As stated in changelog for version 3.0.0, firms switched to reporting accounts receivable in `line_1240` (previously: `line_1230`). Ironically, having warned the users about this we fell into the trap ourselves: our totals adjustment procedure did not respect the switch resulting in incorrect values in `line_1600` and breaking articulation for many firms. 
-- Fixed an incoherity in marking all-zero observations as non-filed and non-imputed. The logic was first introduced in version 2.0.0, but its realization was faulty. Our financial data come from two sources, Rosstat and GIR BO. In the former, all missings are repesented by zeros, and since it is impossible to tell a true zero from a missing we replace all zeros with `NA`s in the period. In the latter, it is different: the data is parsed from XMLs and we can be sure that zeros have indeed been reported, so we do not replace them withs `NA`s. However, our code did not account for this difference between periods: we marked only all-`NA` cases, and not all-zero.
+- Fixed an incoherity in marking all-zero observations as non-filed and non-imputed. The logic was first introduced in version 2.0.0, but its realization was faulty. Our financial data come from two sources, Rosstat and GIR BO. In the former, all missings are repesented by zeros, and since it is impossible to tell a true zero from a missing we replace all zeros with `NA`s in the period. In the latter, it is different: the data is parsed from XMLs and we can be sure that zeros have indeed been reported, so we do not replace them with `NA`s. However, our code did not account for this difference between periods: we marked only all-`NA` cases, and not all-zero.
 - XML path for `line_3600` was broken while updating the parser for 3.0.0. Things are back to normal now.
 
 ### Changed