CAViaR: Conditional Autoregressive Value at risk by Regression Quantile

Hi, i am trying to compute the Engle & Manganelli´s CAViaR, i downloaded the codes from Manganelli´s web page, however while i´m running it i get an error, which says:
Error in ==> RQobjectiveFunction at 39 VaR = SAVloop(THETA, BETA, y, VaR(1)); % Call the C program to compute the VaR loop.
Error in ==> CAViaROptimisation at 119 RQfval(i) = RQobjectiveFunction(initialTargetVectors(i,:), 1, MODEL, inSampleObs, y(:,t), THETA, empiricalQuantile(t));
Error in ==> CAViaR at 38 eval(a)
It is related with C program but the fact is that i don´t know why Manganelli uses such a program to compute the VaR loop. It would be very helpul if someone knows about it.
The codes and the data can be downloaded from: http://www.simonemanganelli.org/Simone/Research.html
Regards. Gustavo

Respuestas (1)

Duminda
Duminda el 22 de Nov. de 2013
Hi,
Those CAViaR codes use C programs associated with MATLAB files. If you run the program in Windows 32bit version, there won't be a problem with C files. Those .DLL files have to be in the same folder. Problems ocurre when you run them in Win 64bit. Because .DLL files are compatible with 32bit. To solve this, you need to recompile these using MATLAB MEX command. To do so, first you need to install Windows SDK 64 bit from http://www.microsoft.com/en-us/download/details.aspx?id=8442. Because MATLAB 64bit does not have a built in C compiler. Further details may be found in http://www.mathworks.com/support/compilers/R2012a/win64.html.
Good luck.

1 comentario

Could you please tell me how to solve the problem when run the code with Matlab 2017b? ( Microsoft Windows SDK 7.1 is not supported as of R2017b)
Many thanks!

Iniciar sesión para comentar.

Categorías

Más información sobre MATLAB Compiler en Centro de ayuda y File Exchange.

Preguntada:

el 19 de Jun. de 2013

Comentada:

el 3 de Jun. de 2018

Community Treasure Hunt

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

Start Hunting!

Translated by