9  Uniform shocks

9.1 Overview

ems_uniform_shock() applies a homogeneous percentage change over an entire variable or a subset of its elements. The set arguments accepted in ... depend on the variable specified and the set mappings loaded in ems_data().

ems_uniform_shock(var, value, ...)

Argument	Type	Description
var	Character	Variable name as it appears in the model file
value	Numeric	Percentage change applied to all targeted elements
...	Named arguments	Set-index pairs restricting the shock. Each value is a single element ("element"), multiple elements (c("a", "b")), or a named subset ("SUBSET")

9.2 Full variable uniform shocks

The simplest uniform shock is applied to a full variable, meaning that all combinations of sets associated with that variable are allocated the same exogenous value. In the example below, all regions (as determined by selected region mappings) will be increased by 1%.

pop_shk <- ems_uniform_shock(var = "pop",
                             value = 1
)

9.3 Partial variable uniform shocks

As with the syntax regarding swaps, the scope of a uniform shock may be determined by specifying elements within the sets associated with the shocked variable. For example, if we have a region “chn” and wish to allocate a shock specifically to this region:

partial <- ems_uniform_shock(var = "aoall",
                             value = -1,
                             REGr = "chn"
)

If a full swap is followed by a partial variable shock, the exogenous unshocked elements will remain at 0.

In the case of intertemporal models, the shock year may be specified using a chronological or timestep format (see “Time step summary” output from ems_data()).

partial <- ems_uniform_shock(
  var = "aoall",
  value = -1,
  REGr = "chn",
  ACTSa = "crops",
  Year = 2017
)

Note that sets specified in this manner are identified by the model-specific concatenation of the standard set name plus the variable-specific index. Also note that exogenous components of a variable not allocated a shock (all other regions except “chn”) will not vary.

9.4 Subset shocks

A named model subset may be passed in place of individual elements, identical to the subset form in ems_swap(). teems expands the subset to all its elements at the aggregation level specified in ems_data().

In the GTAP-RE model, MARG is a subset of COMM containing margin commodities, and FWDTIME is a subset of ALLTIME containing forward time steps. Using these subset names directly restricts the shock to only those elements:

partial <- ems_uniform_shock(
  var = "qxs",
  value = -1,
  COMMc = "MARG",
  REGs = c("chn", "usa"),
  ALLTIMEt = "FWDTIME"
)

As in the swaps example, REGs = c("chn", "usa") is a multi-element specification while COMMc = "MARG" and ALLTIMEt = "FWDTIME" are subset specifications — both forms can be combined freely in a single call. This shock would then be deployed alongside the corresponding subset swaps:

cmf_path <- ems_deploy(
  .data = dat,
  model = model,
  shock = partial,
  swap_in = qxs_in,
  swap_out = txs_out
)

9.5 Multiple shocks and mixed swaps

Multiple uniform shocks of different scope can be combined in a list and passed to ems_deploy(). Swaps can also be mixed, combining partial ems_swap() objects with full variable strings:

partial <- ems_uniform_shock(
  var = "qfd",
  value = -1,
  REGs = "usa",
  PROD_COMMj = "crops"
)

full <- ems_uniform_shock(
  var = "yp",
  value = 0.1
)

qfd <- ems_swap(
  var = "qfd",
  REGs = "usa",
  PROD_COMMj = "crops"
)

tfd <- ems_swap(
  var = "tfd",
  REGr = "usa",
  PROD_COMMj = "crops"
)

cmf_path <- ems_deploy(
  .data = dat,
  model = model,
  shock = list(partial, full),
  swap_in = list(qfd, "yp"),
  swap_out = list(tfd, "dppriv")
)