garchset / garchget to new functions

6 次查看（过去 30 天）
art.anna
art.anna 2017-8-20
回答： Elizabeth Reese 2017-8-23
Good day,
I am trying to adapt the old functions from Econemtrics Toolbox of 2013/14 to the new one. Even though I found the following link with some instructions ( https://uk.mathworks.com/help/econ/converting-from-garch-functions-to-models.html ) I can't manage to adapt it to my case.
I am estimating GJR(1,1). The idea is that the code needs to add volatility shock to the previously estimated model (that is why it uses garchget and multiplies the volatility constant by a shock).
for i=1:nEquities
spec(i)=garchset(spec(i),'k',garchget(spec(i), 'k')* VolaShock(i,1)^2); % apply stress input to GARCH volatility constants
end
An then it simulates the output using garchsim:
for i = 1:nEquities
[~, ~, ...
StressedsimulatedReturns(:,:,i)] = garchsim(spec(i), horizon, nPaths, Z(:,:,i), ...
[], [], preResidual(i), preSigma(i), ...
preReturn(i));
end
I understand that now the functions to use are garch / estimate / simulate but I don't understand how to adapt the syntax... Any help / advice is appreciated.

请先登录，再回答此问题。

回答（1 个）

Elizabeth Reese
Elizabeth Reese 2017-8-23
In order to add all of the parameters you are using, you can consider making a GJR object using the "gjr" function, and then specify that as the "Variance" property in the "arima" constructor and estimate the model. You can set the 'K' value based on the third row of the chart you mentioned. Then, you can simulate the ARIMA model using the Name-Value pairings that map to the additional inputs you were using for "garchsim".

类别

在 Help Center 和 File Exchange 中查找有关 Conditional Variance Models 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by