Sets and Parameters

WiNDCNational is built using WiNDCContainer, which provides a flexible way to define and manage sets and parameters within the data model. We will provide a listing of the sets used and demonstrate how to access and view them in code.

The sets and parameters are the same for both the summary and detailed tables. For this reason we will use the summary table as the basis for our examples.

using WiNDCNational
using DataFrames
summary = build_us_table(:summary);

Sets

To view all the sets, use the sets function:

sets(summary) |> x-> sort(x, [:domain, :name])

40×3 DataFrame

Row	name	description	domain
	Symbol	String	Symbol
1	duty	Duty	col
2	export	Exports	col
3	government_final_demand	Government portions of final demand	col
4	import	Imports	col
5	investment_final_demand	Investment portions of final demand	col
6	margin	Margin sectors	col
7	personal_consumption	Personal consumption expenditures	col
8	sector	Sectors	col
9	subsidy	Subsidy	col
10	tax	Tax	col
11	trade	Trade	col
12	transport	Transport	col
13	Capital_Demand		parameter
14	Duty		parameter
15	Export		parameter
16	Final_Demand	Final demand	parameter
17	Government_Final_Demand		parameter
18	Household_Supply	Positive values in PCE (negative in USE table)	parameter
19	Import		parameter
20	Intermediate_Demand		parameter
21	Intermediate_Supply		parameter
22	Investment_Final_Demand		parameter
23	Labor_Demand		parameter
24	Margin_Demand	Positive values in marginal categories	parameter
25	Margin_Supply	Negative values in marginal categories	parameter
26	Other_Final_Demand	Non-export components of final demand	parameter
27	Output_Tax		parameter
28	Personal_Consumption	Negative values in PCE (positive in USE table)	parameter
29	Sector_Subsidy		parameter
30	Subsidy		parameter
31	Supply	Supply (or output) sections of the IO table	parameter
32	Tax		parameter
33	Use	Use (or input) sections of the IO table	parameter
34	Value_Added	Value added	parameter
35	capital_demand	Gross operating surplus	row
36	commodity	Commodities	row
37	labor_demand	Compensation of employees	row
38	output_tax	Output tax	row
39	sector_subsidy	Sector subsidy	row
40	year		year

The |> x -> sort(x, :[:domain, :name]) sorts the sets by their domain, or the column in the data table where the set elements are located, and their name.

By convention we use snake_case for sets that do not have the parameter domain.

Let's consider the set margin with domain col. I would like to view the elements:

elements(summary, :margin)

2×3 DataFrame

Row	name	description	set
	Any	Any	Symbol
1	Trade	Trade margins	margin
2	Trans	Transport margins	margin

The values in the name column are the elements of the set, they are NAICS codes as they appear in the BEA input/output tables. To examine the actual data use:

table(summary, :margin) |> x -> first(x, 5)

5×5 DataFrame

Row	row	col	year	parameter	value
	Any	Symbol?	Int64	Symbol	Float64
1	111CA	Trans	1998	margin_demand	20.257
2	113FF	Trans	1998	margin_demand	2.421
3	211	Trans	1998	margin_demand	22.533
4	212	Trans	1998	margin_demand	14.247
5	321	Trans	1998	margin_demand	4.592

This will show the first 5 rows of the data table restricted to the margin set. Recall margin had domain col and that is the column where the elements live. To demonstrate the multiple values in col, just view the unique values:

table(summary, :margin) |> x -> unique(x, :col)

2×5 DataFrame

Row	row	col	year	parameter	value
	Any	Symbol?	Int64	Symbol	Float64
1	111CA	Trans	1998	margin_demand	20.257
2	111CA	Trade	1998	margin_demand	45.523

Because we are only viewing the first few rows, we only see the Trans code. We can further filter the data by specifying a particular element:

table(summary, :margin => :Trade) |> x -> first(x, 5)

5×5 DataFrame

Row	row	col	year	parameter	value
	Any	Symbol?	Int64	Symbol	Float64
1	111CA	Trade	1998	margin_demand	45.523
2	113FF	Trade	1998	margin_demand	5.41
3	211	Trade	1998	margin_demand	3.609
4	212	Trade	1998	margin_demand	2.508
5	321	Trade	1998	margin_demand	17.94

Again, we can extract the unique values:

table(summary, :margin => :Trade) |> x -> unique(x, :col)

1×5 DataFrame

Row	row	col	year	parameter	value
	Any	Symbol?	Int64	Symbol	Float64
1	111CA	Trade	1998	margin_demand	45.523

Parameters

Parameters are just sets with domain :parameter.

sets(summary) |> x-> subset(x, :domain => ByRow(==(:parameter)))

22×3 DataFrame

Row	name	description	domain
	Symbol	String	Symbol
1	Export		parameter
2	Subsidy		parameter
3	Labor_Demand		parameter
4	Capital_Demand		parameter
5	Investment_Final_Demand		parameter
6	Import		parameter
7	Sector_Subsidy		parameter
8	Government_Final_Demand		parameter
9	Duty		parameter
10	Intermediate_Demand		parameter
11	Intermediate_Supply		parameter
12	Tax		parameter
13	Output_Tax		parameter
14	Value_Added	Value added	parameter
15	Final_Demand	Final demand	parameter
16	Other_Final_Demand	Non-export components of final demand	parameter
17	Use	Use (or input) sections of the IO table	parameter
18	Supply	Supply (or output) sections of the IO table	parameter
19	Margin_Demand	Positive values in marginal categories	parameter
20	Margin_Supply	Negative values in marginal categories	parameter
21	Household_Supply	Positive values in PCE (negative in USE table)	parameter
22	Personal_Consumption	Negative values in PCE (positive in USE table)	parameter

By convention, parameter names are capital-version of snake_case. Sets are required to have distinct names and this avoid conflicts.

Extracting parameters is identical to extracting sets.

table(summary, :Intermediate_Demand) |> x -> first(x, 5)

5×5 DataFrame

Row	row	col	year	parameter	value
	Any	Symbol?	Int64	Symbol	Float64
1	111CA	111CA	1998	intermediate_demand	-33.87
2	113FF	111CA	1998	intermediate_demand	-12.047
3	212	111CA	1998	intermediate_demand	-1.67
4	22	111CA	1998	intermediate_demand	-3.378
5	23	111CA	1998	intermediate_demand	-0.796

We can extract multiple parameters:

table(summary, :Intermediate_Demand, :Intermediate_Supply) |> x -> first(x, 5)

5×5 DataFrame

Row	row	col	year	parameter	value
	Any	Symbol?	Int64	Symbol	Float64
1	111CA	111CA	1998	intermediate_demand	-33.87
2	113FF	111CA	1998	intermediate_demand	-12.047
3	212	111CA	1998	intermediate_demand	-1.67
4	22	111CA	1998	intermediate_demand	-3.378
5	23	111CA	1998	intermediate_demand	-0.796

Some parameters are composites of multiple parameters:

elements(summary, :Value_Added)

2×3 DataFrame

Row	name	description	set
	Any	Any	Symbol
1	labor_demand	Value added	Value_Added
2	capital_demand	Value added	Value_Added

Composite Parameters

Some parameters are require aggregations and can't be extracted directly. Here is a full list of these parameters:

This page was generated using Literate.jl.