Hello blog, we haven’t seen each other for a while – I’ve been meaning to call, but you know how it is, just been so busy with work and all that…
What do you mean I only visit when I want to use you for something? That’s not fair, I do check up on you now and again, although you’ve been looking a little forlorn – like you’ve not been looking after yourself properly – so I’ve kept my distance.
I’ve brought you a present though! I brand-spanking new re-hash of some work I did last year, but that’s been sitting on my hard drive waiting to be cobbled together into something approaching a paper…
The lovely folks at CASA have honoured it with working paper status, so if anyone wants a step-by-step guide of how to run a spatial interaction model in R, then this is the place to visit:
I know you like pictures, so before I go here’s a tasty little number from that very paper
Just to prove that I’ve not been twiddling my thumbs for the last 12 months, here’s the first (of hopefully many) working papers to come out of the ENFOLDing Migration stable…
Alan Wilson and I have been working on a new family of Spatial Interaction Models for estimating inter-regional migration flows in Europe. We’ve developed a new multi-level model which uses internal migration data at the regional level and international migration data at the country level to model international regional level flows.
Once I’ve finished tidying up the code, I’ll post the full model here too for people to play with – at the moment though, you’ll just have to make do with the paper, which can be found here:
And here’s a pretty picture from it!
Yes, British migration classification fans, it’s the moment you’ve all been waiting for – publication of the CIDER Migration Classification paper! *Cue delirious cheering, whooping, hollering and cries of ‘get in the hole!!’*
This most recent product of my blood, sweat and tears – co-authored by John Stillwell – can be found in the latest edition of Population Trends. I say latest, but I should also say final, as sadly after 36 years of publication Population Trends is from today, no more. In 2010 the Office for National Statistics took the decision to cease regular publication of all of its journal titles – I can only speculate the victim of the severe cuts being enforced across the public sector currently.
The Autumn 2011 issue really does mark the end of an era and I will be sad to see it go, especially as I think I can safely say that it is the journal, above all others, that has been the source of useful and practical research during my academic career.
Iterative Proportional Fitting? That old chestnut? Haven’t people been banging on about that for hundreds of years?! Well, about 70, but I guess some of the old ones are the best ones… (Deming and Stephan for all those pub quiz fans out there!)
Anyway, I recently stumbled across IPF again in the way that people who work with data matrices do from time to time. I also realised that there’s not a whole lot of useful generic programmes out there for you to carry out this pretty useful procedure, and like a total nit, I’ve been doing it the hard way for a long time. Therefore I have spent the last day or so writing a nice little bit of VBA (arrghh, not Microsoft?! Yes! And it even runs in the much maligned but fantastic Excel). The program will take in data in either matrix or paired list format and will run IPF to update the interior cell values to a new set of column and row margins whilst maintaining the original ratio structure of the table.
I think Paul Norman produced a similar program in Excel a few years ago, but if you can find it anywhere on the web you’re a better Googler than me! Hopefully this programme improves Paul’s a little with the flexibility for matrices and paired lists and with it taking advantage of the increased number of columns available in the latest versions of Excel. Also, this will be freely available to download from here from now on. If you’d like to use the program, it can be downloaded from the link below:
Oh and here’s a nice picture that my ENFOLD-ing colleagues will enjoy!