BookmarkSubscribeRSS Feed
BingL
Obsidian | Level 7

Hello to whoever can help

 

please review the following program

proc autoreg data=exam4_1;
	model Rtn = / method=ml archtest
				garch=(p=1, q=1);
	output out=FTSE100 r=ehat_garch ht=hgarch;
	title "Estimate GARCH(1,1) Model and Forecast Volatility";
run;

I kinda doult that the moethd=ml is using the Levenberg-Marquardt Algorithm, because the result numbers are not close from those of my book.

 

If not, how to use Levenberg-Marquardt method in PROC AUTOREG.

 

Thank you.

1 REPLY 1
PGStats
Opal | Level 21

The optimisation method will determine how quickly you get to the solution but should not change the solution. The difference is more likely due to the type= of GARCH model that is fitted (NELSONCAO by default). 

PG

hackathon24-white-horiz.png

The 2025 SAS Hackathon has begun!

It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.

Latest Updates

Discussion stats
  • 1 reply
  • 1456 views
  • 0 likes
  • 2 in conversation