Third time’s a charm

After getting rejected by an applied statistics journal and a more methodological statistics journal we’ve decided to send the Finnish paper to an environmental statistics journal. I think it’s a more natural home for this paper as it straddles the boundary of model development and the analysis of observational environmental data. A new version of the preprint will appear on arXiv in about 12 hours.

I’m also still redrafting the first UPTECH paper of my thesis. With the Finnish paper resubmitted it’ll be the focus of my work over the next few days. I think the modelling is now much improved (although I might try an alternative specification) after taking reviewer comments into account from its earlier rejection. Once I’ve got this last modelling question sorted out and the analysis is written up (not much has to change in the methodology section) I can restart work on my final thesis paper.

There are two similar papers coming up in my immediate post-thesis future where I’m going to have the chance to learn more about the use of Gaussian processes. While INLA does some nice GMRF approximations for Gaussian fields [1] there are some challenges that we’ll face with this data. I might also need to read up on the use of dynamic linear models for modelling the evolution of a parameter with time.

[1] Lindgren, F.; Rue, H. & Lindström, J. (2011) An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach, Journal of the Royal Statistical Society: Series B (Statistical Methodology) 73, 423-498. R-INLA.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s