Subsetting the EU data

[wengraf]

@Robinlovelace: We need to work out which vehicles we are keeping. I would suggest we need all GB, plus all non-GB that isn't replicated by GB equivalents.

[wengraf]

@Robinlovelace: I've added some elements and tidied up the data - there is new registration data, which should be the basis of a joining variable. The commercial model name also looks promising in format. Proposed next steps:

Switch to reading final data. For example:

• 2018: https://cmshare.eea.europa.eu/s/Wme8dMPtS8LmWgk/download
• 2017: https://www.eea.europa.eu/data-and-maps/data/co2-cars-emission-18/co2-emissions-cars-2017-final/co2_passengers_cars_v16_csv.zip/at_download/file
• 2016: https://www.eea.europa.eu/data-and-maps/data/co2-cars-emission-18/co2-emissions-cars-2016-final/co2_passenger_cars_v14_csv.zip/at_download/file

Read in VEH160 from the DfT Vehicle Stats. VEH160 has DVLA model name and total new registrations. GB subset of EU data has "commercial model name", which looks quite DVLA-ish at first glance, and total new registrations, which ought to match UK annual new registration totals.

Test how much of the new registration UK vehicle parc is submitted to EU.

Are you getting full or generic model name in your stats19 data? I think you really need full.