Tag Archives: algebra

An Example of MBH "Robustness"

Apr 5, 2008 – 3:15 PM

In the MBH source code, they apply steps that purport to weight the temperature PCs in their regression calculations proportional to their eigenvalues. Comments on their code say: c set specified weights on data … c weights on PCs are proportional to their singular values This is one of two weighting procedures in MBH for […]

By Stephen McIntyre | Posted in MBH98, Replication, Source Code | Also tagged , | Comments (30)

Squared Weights in MBH98

Apr 5, 2008 – 11:39 AM

A couple of weeks ago, I said that I would document (at least for Jean S and UC) an observation about the use of squared weights in MBH98. I realize that most readers won’t be fascinated with this particular exposition, but indulge us a little since this sort of entry is actually a very useful […]

By Stephen McIntyre | Posted in MBH98, Multivariate, Replication, Source Code | Also tagged , | Comments (15)

MBH and Partial Least Squares

Jun 13, 2006 – 8:35 AM

Right now, I’m working on two main projects where I intend to produce papers for journals: one is on the non-robustness of the “other” HS studies; the other is on MBH98 multivariate methods. The latter topic is somewhat “in the news” with the two Bürger articles and with the exchange at Science between VZ and […]

By Stephen McIntyre | Posted in MBH98, Multivariate, Replication, Statistics | Also tagged , , , , , , , | Comments (41)

More Linear Algebra

Feb 13, 2006 – 4:08 PM

I’m not sure that there’s a huge demand for more linear algebra on MBH98, but here’s the rest of the proof that the NH temperature index in an MBH98-type calculation is simply a linear combination of proxies and, when only one temperature PC is reconstructed, the weights are proportional to the correlation between each proxy […]

By Stephen McIntyre | Posted in MBH98, Replication, Source Code | Also tagged , | Comments (16)

MBH Calibration-Estimation Procedure

Feb 6, 2006 – 10:46 AM

What I’m going to show here is that the MBH98 method can be reduced to a few lines of code and, in doing so, show some other interesting results as well. Today I’m just going to get to the reconstructed temperature PCs, but I’ll show that these are linear in the proxies and later show […]

By Stephen McIntyre | Posted in MBH98, Replication, Source Code | Also tagged | Comments (47)