2017 Supply and Use tables from useeior

[WesIngwersen]

• Adding 2017 Detail and 2017-2022 Summary Supply and Use tables and updating script notes. Manually spot-checked csv table data against originals.
• Enables use of 2017 NAICS crosswalk and target_naics_year
• Adds FBS of annual summary use table attributed to detail sectors proportionally to Detail 2017 Use table

[WesIngwersen]

fyi @catherinebirney note that I pulled in and added the metadata json files from useeior for these two files... i found that would be very valuable to have here

[catherinebirney]

fyi @catherinebirney note that I pulled in and added the metadata json files from useeior for these two files... i found that would be very valuable to have here

I have wondered if it would be worth putting the csv files (and metadata) on the useeior data commons repo instead of within flowsa and have flowsa load the data from there - it could be a cleaner/easier way of tracking what files we use

[bl-young]

I have wondered if it would be worth putting the csv files (and metadata) on the useeior data commons repo instead of within flowsa and have flowsa load the data from there - it could be a cleaner/easier way of tracking what files we use

I think that's a nice idea. Since they are rda files they won't be impacted by the useeior package after they are created, so just using the metadata from useeior (as Wes did here) and saving them on data commons would be sufficient. We would not need these scripts anymore to document them.

[bl-young]

Also I believe there is some conflict/duplication here with edits made to the scripts. Now there is both a pull_BEA_Make_and_Use_csv.py and write_BEA_data_from_useeior.py

[catherinebirney]

Also I believe there is some conflict/duplication here with edits made to the scripts. Now there is both a pull_BEA_Make_and_Use_csv.py and write_BEA_data_from_useeior.py

you can keep 'write_BEA_data_from_useeior' and delete 'pull_BEA_Make_and_Use_csv.py' - I was just trying to clean up the documentation yesterday

[bl-young]

@catherinebirney do you want to work on the FBAs (and FBS?) here in this branch?

[catherinebirney]

@catherinebirney do you want to work on the FBAs (and FBS?) here in this branch?

@bl-young Sure - I will work on the FBAs now in this branch.

[catherinebirney]

@bl-young I added the initial FBA code - I'm checking for any issues now. Do you want to generate the FBS?

[bl-young]

@bl-young I added the initial FBA code - I'm checking for any issues now. Do you want to generate the FBS?

Sure i will give it a shot

[bl-young]

@catherinebirney draft BEA crosswalk added. It needs more review and will likely change, but includes 2012 and 2017 NAICS for now.

[bl-young]

@catherinebirney I've reviewed the use tables, including converting them back to summary using map_to_BEA_sectors(), while they aren't a perfect match (due to many:many NAICS:BEA), I feel comfortable moving forward. Any thing left on your end before we pull this into the other branch?

[catherinebirney]

@catherinebirney I've reviewed the use tables, including converting them back to summary using map_to_BEA_sectors(), while they aren't a perfect match (due to many:many NAICS:BEA), I feel comfortable moving forward. Any thing left on your end before we pull this into the other branch?

The Detail_Use_year tables do not get their NAICS codes converted to target year 2012- they are remaining in NAICS 2017. I'm working on addressing why Nevermind - the conversion happens later in the method. Let me finish running this FBS and then I am fine with pulling this into the state ghg branch. I'll comment again

[bl-young]

~The Detail_Use_year tables do not get their NAICS codes converted to target year 2012- they are remaining in NAICS 2017. I'm working on addressing why~ Nevermind - the conversion happens later in the method. Let me finish running this FBS and then I am fine with pulling this into the state ghg branch. I'll comment again

Yeah it looks like its working fine in the CAP HAP methods

[catherinebirney]

@bl-young The BEA_Use tables naics are now being converted correctly - I agree this PR can be merged