RE: Slow Gradient Method.

From: Vladimir Piotrovskij Date: May 30, 2001 technical Source: cognigencorp.com
From: "Piotrovskij, Vladimir [JanBe]" <VPIOTROV@janbe.jnj.com> Subject: RE: Slow Gradient Method. Date: Wed, 30 May 2001 14:18:25 +0200 Thanks Bill for your explanation. Now I understand that SLOW can hardly have any advantages over the default algorithm. Meanwhile I performed some exploration and found that using SLOW with METHOD=COND did not improve the convergence behaviour in complicated cases. I have currently one example where FOCE fails to converge whereas FO method converges perfectly. With FOCE NONMEM stops due to rounding errors, and before this it does a few iterations with very high gradient (>10^5) and no changes in parameters at all. Best regards, Vladimir

Thread

View all 12 messages
May 24, 2001 Matt Hutmacher Slow Gradient Method.
May 25, 2001 Niclas Jonsson Re: Slow Gradient Method.
May 28, 2001 Vladimir Piotrovskij RE: Slow Gradient Method.
May 28, 2001 Nick Holford Re: Slow Gradient Method.
May 30, 2001 William Bachman RE: Slow Gradient Method.
May 30, 2001 Vladimir Piotrovskij RE: Slow Gradient Method.
May 30, 2001 Stephen Duffull RE: Slow Gradient Method.
May 31, 2001 Niclas Jonsson RE: Slow Gradient Method.
May 31, 2001 Erik Olofsen RE: Slow Gradient Method.
May 31, 2001 Nick Holford Re: Slow Gradient Method.
May 31, 2001 Niclas Jonsson Re: Slow Gradient Method.
May 31, 2001 Stuart Slow Gradient Method
← Previous
Next →