| compute_score {globaltrends} | R Documentation |
Compute keyword-country search score
Description
The function computes search scores for object keywords. Search volumes for
control and object batches are mapped to the same base. Next, search volumes
for object batches are divided by the sum of search volumes for the
respective control batch. compute_voi computes volume of
internationalization (VOI) as global search scores.
Usage
compute_score(object, control = 1, locations = gt.env$countries)
## S3 method for class 'numeric'
compute_score(object, control = 1, locations = gt.env$countries)
## S3 method for class 'list'
compute_score(object, control = 1, locations = gt.env$countries)
compute_voi(object, control = 1)
Arguments
object |
Object batch for which the data is downloaded. Object
of type |
control |
Control batch for which the data is downloaded. Object
of type |
locations |
List of countries or regions for which the data is
downloaded. Refers to lists generated in |
Details
The search score computation proceeds in four steps. First, the function
aggregates all search volumes to monthly data. Then, it applies some optional
time series adjustments: seasonally adjusted forecast::seasadj and
trend only stats::stl. Next, it follows the procedure outlined by
Castelnuovo and Tran (2017, pp. A1-A2) to map control and object data. After
the mapping, object search volumes are divided by the sum of control search
volumes in the respective control batch. We use the sum of search volumes for
a set of control keywords, rather than the search volumes for a single
control keyword, to smooth-out variation in the underlying control data.
Castelnuovo, E. & Tran, T. D. 2017. Google It Up! A Google Trends-based Uncertainty index for the United States and Australia. Economics Letters, 161: 149-153.
Value
Message that data has been computed successfully. Data is written to table data_score.
Note
When synonyms were specified through add_synonym, search
scores for synonyms are added to the main keyword.
See Also
Examples
## Not run:
compute_score(
object = 1,
control = 1,
locations = countries
)
compute_voi(
object = 1,
control = 1
)
compute_score(
object = as.list(1:5),
control = 1,
locations = countries
)
## End(Not run)