Simulate multivariate normal asset return scenarios from data
Using a fints
object for the
AssetReturns
argument of
simulateNormalScenariosByData
is not recommended. Use timetable
instead for financial time series. For more information, see
Convert Financial Time Series Objects fints to Timetables.
simulates multivariate normal asset return scenarios from data for portfolio
object for obj
= simulateNormalScenariosByData(obj
,AssetReturns
)PortfolioCVaR
or PortfolioMAD
objects. For details on the workflows, see PortfolioCVaR Object Workflow,
and PortfolioMAD Object Workflow.
simulates multivariate normal asset return scenarios from data for portfolio
object for obj
= simulateNormalScenariosByData(obj
,AssetReturns
,NumScenarios
,Name,Value
)PortfolioCVaR
or PortfolioMAD
objects using additional options specified by one or more
Name,Value
pair arguments.
This function estimates the mean and covariance of asset returns from either
price or return data and then uses these estimates to generate the specified
number of scenarios with the function mvnrnd
.
Data can be in a NumSamples
-by-NumAssets
matrix of NumSamples
prices or returns at a given periodicity
for a collection of NumAssets
assets, a table
or a timetable
.
Note
If you want to use the method multiple times and you want to simulate
identical scenarios each time the function is called, precede each
function call with rng
(seed) using a specified integer
seed.
You can also use dot notation to simulate multivariate normal asset return scenarios from data for a PortfolioCVaR or PortfolioMAD object.
obj = obj.simulateNormalScenariosByData(AssetReturns,NumScenarios,Name,Value);