🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: sci.stat.consult
Article #3

How to calculate the hessian???

#3
From: Daan Taks
Date: Mon, 21 Jan 2002 13:49
11 lines
1662 bytes
In Excel I calculated a variance-covariance (Ht) (2*2) matrix using
the ADC model of Kroner and Ng (1998). Now I want to calculate the
standard errors of the parameters (16 in total) estimated by the
model. I know that I can do that by calculating the hessian, but how
to do that???
The log-likelihood function is as follows:
-0.5*log(determinant of Ht) - 0.5*E'*(inverse of Ht)* E
Where E is an 2*1 vector with return1 and return2 of stock indices.
How can I differentiate this log-likelihood function???
Help would be very appreciated!!!!!
Greetings Daan.
Technical Details
Message-ID: <3172868c.0201210549.66390229@posting.google.com>
Path: arc1-ca.pugleaf.net!archiver1.google.com!postnews1.google.com!not-for-mail

Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 106e3f,74636f6e43f5c32c,start
X-Google-Attributes: gid106e3f,public
X-Google-Thread: 10d632,74636f6e43f5c32c,start
X-Google-Attributes: gid10d632,public
X-Google-Thread: 111b91,74636f6e43f5c32c,start
X-Google-Attributes: gid111b91,public
X-Google-Thread: fa931,74636f6e43f5c32c,start
X-Google-Attributes: gidfa931,public
X-Google-ArrivalTime: 2002-01-21 05:49:17 PST
Path: arc1-ca.pugleaf.net!archiver1.google.com!postnews1.google.com!not-for-mail
From: blackfiction@hotmail.com (Daan Taks)
Newsgroups: sci.stat.consult,sci.stat.math,comp.soft-sys.matlab,aus.stats.s
Subject: How to calculate the hessian???
Date: 21 Jan 2002 05:49:16 -0800
Organization: http://groups.google.com/
Lines: 11
Message-ID: <3172868c.0201210549.66390229@posting.google.com>
NNTP-Posting-Host: 137.120.186.171
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1011620957 8660 127.0.0.1 (21 Jan 2002 13:49:17 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: 21 Jan 2002 13:49:17 GMT