documentation for 26 file CGE download

I have a question about what I think are institutional imports. Totals from what I think are the foreign and domestic institutional imports (7 by 4 and 8 by 4?) seem to differ from totals in the SAM. Why would they differ and should they differ?

PS I'm looking at Clark Co. NV 2010.

Hello, This SAM contains full detail, including imports and exports. This report generates 26 text files (of the form *1x1.dat) which are inputs into the GAMSTM software for computable general equilibrium models (CGEs). More information can be found here: https://implan.com/v4/index.php?option=com_docman&task=doc_download&gid=225&Itemid=7 Let us know if you have any further questions. Thanks

When I look at the 5 by 4 block totals of foreign imports for each institution and compare those with totals from the disaggregated version in 7 by 4 they are not the same. At least that's what I seem to be getting Similarly for the 6 by 4 and 8 by 4 versions. Should these totals be the same?

[quote="E. Fadali" post=14078]I have a question about what I think are institutional imports. Totals from what I think are the foreign and domestic institutional imports (7 by 4 and 8 by 4?) seem to differ from totals in the SAM. Why would they differ and should they differ?[/quote] I aggregated the US 2010 data to 3 sectors, and printed the 26 CGE files. In Excel I did a pivot table and compared to the IMPLAN reports as noted below. 7x4 Institutional foreign import use - foreign imports to institutional demand by commodity 8x4 Institutional domestic import use - did not have values in US data (which makes sense as there are no domestic imports) 2x4 Domestic institutional use or final demands by institution - US institutional use of US produced commodities I compared 7x4 plus 8x4 to the Social Accounts > Export > Structural Matrix Elements (CSV Only) > Institution Imports reports: all data elements were the same I compared the 2x4 to the Social Accounts > Social Accounts Reports > View by: Institution Local Commodity Demand (for 14001, 14002): after converting to millions values were the same I compared the 2x4 to the Social Accounts > Social Accounts Reports > View by: Household Local Commodity Demand (for 10001:10009): after converting to millions values were the same I compared the 2x4 to the Social Accounts > Social Accounts Reports > View by: Government Local Commodity Demand (for 11001:11003 & 12001:12003): after converting to millions values were the same Finally I summed 2x4 with 7x4 plus 8x4 (empty) and compared to Study Area Data reports..... Sum to the Study Area Data > View by: Institution Commodity Demand (for 14001, 14002): after converting to millions values were the same I compared the 2x4 to the Study Area Data > View by: Household Commodity Demand (for 10001:10009): after converting to millions values were the same I compared the 2x4 to the Study Area Data > View by: Government Commodity Demand (for 11001:11003 & 12001:12003): after converting to millions values were the same If the problem lies in the domestic imports (8x4), this would not have caught it, but I suggest you test by adding 7x4 with 8x4 and compare to the report as I did above.

The U.S. will likely not have the same problem because it only has one kind of import: foreign. The issue I'm having has to do with the subtotals for foreign and domestic being different. For example: In 5 by 4 Clark 2010 model: 25001.10001 242.8396759 In 7 by 4 Clark 2010 model after summing over all commodities for households 10001: 86.02131113 Foreign plus domestic imports total IS the same (again using household category 10001): 86.02131113 (from 7 by 4 total) + 451.9821618 (from 8 by 4 total) = 538.0034729 242.8396759 (25001.10001 from 5 by 4) + 295.1637878 (28001.10001 from 6 by 4) = 538.0034637 I have modified this model. Could that have created a problem? I will look at the unmodified 26 CGE files.

A quick check of an unmodified Nevada state model 26 CGE files seems to show the same type of mismatch.

The problem seems to be that the 7x4 (foreign commodity import) sums for each institution does not seem to match the 5x4 which is also supposed to be the sum of the foreign import of commodities to institutions. The 7x4 is correct, in that 7x4 plus 8x4 plus 2x4 = total institutional demand for the commodity as shown in the IMPLAN reports. The 5x4 matches the IMPLAN SAM reports but does not match the 7x4, which makes me believe that the 5x4 includes a balancing item, which surprises me because I thought households were balanced through the Capital (savings/dis-savings). I will have to check with Scot when he gets back. Doug

OK I'll stay tuned then. I wondered if there was some balancing going on.

Any news about this? I am thinking I could have the foreign and domestic trade sums stay in the same proportions as the 5 by 4 file. The 5 by 4 data is what I see in the SAM in IMPLAN.

Sorry, yes the difference is balancing. The proportions will work. We are looking into a complete redo of the SAM process to minimize the balancing necessary at a local level, but that doesn't help the current situation.