The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
SWIM 1.0.0 -
current develop version on GitHub
Major changes:
Additional functions and features
Wasserstein distance
stress_wass()
:
- A wrapper for the stress functions using the 2-Wasserstein
distance
stress_RM_w()
:
- a stressed model component (random variable) fulfills a constraint
on its risk measure defined by a gamma function.
stress_RM_mean_sd_w()
:
- a stressed model component (random variable) fulfills a constraint
on its mean, standard deviation, and risk measure defined by a gamma
function.
stress_HARA_RM_w()
:
- a stressed model component (random variable) fulfills a constraint
on its HARA utility defined by a, b and eta parameter and risk measure
defined by a gamma function.
stress_mean_sd_w()
:
- a stressed model component (random variable) fulfills a constraint
on its mean and standard deviation.
stress_mean_w()
:
- a stressed model component (random variable) fulfills a constraint
on its mean.
Functions
mean_stressed()
:
- sample mean of chosen stressed model components, subject to the
calculated scenario weights.
sd_stressed()
:
- sample standard deviation of chosen stressed model components,
subject to the calculated scenario weights.
var_stressed()
:
- sample variance of chosen stressed model components, subject to the
calculated scenario weights.
cor_stressed()
:
- sample correlation coefficient of chosen stressed model components,
subject to the calculated scenario weights.
cdf_stressed()
:
- the empirical distribution function of a stressed model component
(random variable) under the scenario weights.
rename_SWIM()
:
- Get a new SWIM object with desired names.
Features
stress()
:
- A parameter “names” to all stress functions, which allows to name a
stress differently than just “stress 1”, “stress 2”, etc.
- A parameter “log” that allows users to inspect weights’ statistics,
including minimum, maximum, standard deviation, Gini coefficient, and
entropy.
sensitivity()
:
- A parameter “p” can be specified for the degree of Wasserstein
distance.
Minor changes
- fix minor bug in
summary()
.
- add
base
argument for quantile_stressed()
and an error message if the input has wCol
has dimension
larger than 1.
SWIM 0.2.2 - current
version on CRAN
Major
changes: Additional functions and features
plot_quantile()
:
- the function plots the empirical quantile of model components,
subject to scenario weights.
plot_weights()
:
- the function plots the scenario weights of a stressed model against
model components.
stress_moment()
:
- add parameter “normalise” that allows to linearly normalise the
values called by
nleqslv
.
- the function prints a table with the required and achieved moments
and the absolute and relative error.
stress_VaR_ES()
:
- add parameter “normalise” that allows to linearly normalise the
values before
uniroot
is applied.
Minor changes
- fix bug in merging different stress objects.
SWIM 0.2.1
Minor changes
- add vignette
- fix bug in
merge()
.
- fix bug in
sensitivity()
.
SWIM 0.2.0
Major changes
Additional functions and
data sets
VaR_stressed()
:
- the function calculates the VaR of model components, subject to
scenario weights.
ES_stressed()
:
- the function calculates the ES of model components, subject to
scenario weights.
credit_data
:
- a data set containing aggregate losses from a credit portfolio,
generated through a binomial credit model.
Amendments to functions
stress_VaR()
:
- amendment to the calculation of scenario weights when the specified
VaR cannot be achieved.
- returns a message if the achieved VaR is not equal to the stressed
VaR specified.
- specs of the
SWIM
object contains the achieved VaR
- allowing for stressing VaR downwards
stress_VaR_ES()
:
- amendment analogous to the
stress_VaR()
.
- returns a message if the achieved VaR is not equal to the stressed
VaR specified.
- specs of the
SWIM
object contains the achieved VaR
- allowing for stressing VaR and ES downwards
Minor changes
stress()
:
- parameter
x
can have missing column names.
stress_moment()
:
- additional parameter
show
; if TRUE
(default is FALSE
), the result of nleqslv()
is
printed.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.