Today this came up on my Google Scholar alert for 'age-structure' (*)
"Radiostratigraphy and age structure of the Greenland Ice Sheet" (paywalled, I think)
Here is a YouTube video showing age-structure cross-sections of what I suppose to be the same data (not paywalled!)
This is an example of a non-biological aggregate for which we care much more about the thanatological age structure (years left) than the chronological age structure. And it could also be (and probably is, I'm no earth scientist) modeled as a renewal process (winter accumulate, summer thaw), where the longevity of ice sheet 'birth cohorts' will depends on the cumulative net rate of accumulation (or something like that). In like manner, the top-melting over the course of a year is like a thanatological death cohort. Obviously the models needed here are much more complex than this, but it seems like a relevant segue for the Lotka-Euler renewal model both in its chronological and thanatological forms (the later one is a paper I'm soon to submit somewhere).
And here's another video of age-structured ice, this time in the arctic. This one might do better as an age-structured renewal model because depth is less a factor than in Greenland:
*yes, it's a wide net, but I cast it to catch things like this
Since joining the HMD I often get questions, but usually these are special-case questions, not of general interest. This is one is starting to seem frequent. In fact, I've asked that this response make it to the FAQ. That may or may not happen, and in either case, here's my response:
(This is HMD-specific, just to be clear)
Why are many death counts not represented as integers?
There are several reasons why this may happen. (i) Often, deaths of unknown age are distributed (proportionately) over deaths of known age, and this produces decimals. (ii) If input death counts have an open age group, we also distribute this quantity over higher ages using a method outlined in the Methods Protocol. (iii) In some cases, historical infant death counts have been adjusted due to changes in the definition of live births, and these methods may also produce decimals. (iv) In most cases where death counts are represented in Lexis shapes other than that in which they were collected, decimal fractions will result. For instance, 5-year age groups that are split into single ages usually produce decimal estimates in single ages (see Methods Protocol). The same is true of period-cohort counts that are split and combined into age-period counts. (v) Other unique circumstances, such as special data collection procedures (USA) or wartime estimation (ITA, FRA, etc), or the estimation of small but bounded counts (AUS) may also produce decimal death count data. In all cases, even though fractional deaths are impossible, we use them as the best estimate available.
This year's PAA is going to be loads of fun. You can find me there (not necessarily in chronological order):
1) Hosting the session 63 on visualizing demographic data! Four great papers to be presented there, plus discussion from dataviz master, Robert Chung.
2) Poster session 2: "Distributional Aspects of Time to Death in Human Populations". This is a newish idea (for me anyway) that Vladimir Canudas Romo and Adam Lenart are helping me work on and think through clearly (although both may need to ghost-out of the program, due to excessive productivity!).
3) Poster session 4: "Time-to-Death Patterns in Markers of Age and Dependency". I presented in December in Vienna, and am in the middle of writing up a proper paper. There is a lot to be said! Paul Chung (Pil H Chung), Jeroen Spijker (he's going back to the CED soon!) and John MacInnes are also working on this with me. (I'll have to remove my name from the program due to the PAA quota rule, ouchy.)
4) Session 234 on the formal demography of mortality, presenting the paper "Life Lost, Lifesaving, and Causes of Death." Aida Sole i Auro is also working on this with me, and helping me think it through clearly.
5) Session 147 on mortality trends: Magali Barbieri will present the paper "The Cardiovascular Revolution in the United States. A Geographic Analysis", where my role thus far has been in programming. Nadine Oullette (institutionally correct, but this page does no justice!), Mila Andreeva (formerly of CUNY and HMD) and Celeste Winant (HMD) are also involved in diverse aspects of authorship behind this paper. (The program quota also killed me from this one)
6) the EDSD alumni brew-ha-ha, which I'll try to get moving along sometime soon. This is just an announcement that there will be one, I just don't know anything about the logistics of it yet. I'll post it on this site once something is known.
7) At the HFD side-meeting, on April 29, 12:00, demonstrating an easy way to get HFD and HFC data into R.
* and other social gatherings, of course (Berkeley, CED) *
In July, 2015 I'll start as a research scientist at the Max Planck Institute for Demographic Research (MPIDR) in Rostock, Germany! Awesome place, as are all the demography hubs that I've been fortunate enough to stop at and stay a while: CED, Lund, Berkeley. It's time to focus on research-- that's why I'm making the move. I'm grateful to the HMD team and to everyone in the Department of Demography for teaching me so much and making me feel at home the past years. I learned a lot, both from getting my hands dirty with the HMD methods protocol and expanding my bag of formal demographic tricks (these float in the department). Actually it turns out I may not so easily shake off the HMD Methods Protocol, but that's another story. There's always someone to talk with about crazy demography ideas in Berkeley. It's a geek bank that I look forward to visiting regularly-- and getting everything I can out of in the next 6 months! Both Berkeley and Rostock have been in flux the past few years, due to moves and retirements and the like, and it's pleasing to see that both are now stable and growing.
So what's the plan at the MPIDR? I get to work on all my currently-neglected papers. Recent and forthcoming conference presentations are representative of my work, so you could pack it together and call it an agenda. The present thematic unity is coincidental, though, since it's due to a curiosity-driven random walk. So now (err, starting in July) I'll walk with greater purpose and for the sake of walking, rather than walking between tasks. The idea is to get more walking in, and in fact only walk, and hopefully even run. You get the idea.
Why at the MPIDR? Because it has lots of really good people. Lots of them! There are always different people coming and going- they make lots of use of the Guest Researcher funding instruments to bring people in for short and long stays. That's lots of people to pester with questions, good for me. Lots of projects to learn from and give to (This is true of everywhere I've been... I guess sometimes just a change of scenery helps) And it's the hub of the European Doctoral School of Demography (EDSD), exciting, and institutionally close to everywhere that does demography in Europe, including the CED (Ort meiner demografische Erziehung). Ich will/muss sowieso mein Deutsch auffrischen, so there's that. And mostly for the no-strings-attached research freedom.
FYI, there are open calls for several PhD,postdoc, and research scientist jobs at the MPIDR. Look into those!
In case there isn't a happy hour (err, subjective wellbeing seminar) yet at the MPIDR... can you can darn toot'n well count on there being one in the near future. I'll bet they have one, though, what with having a great local brewery. I wonder if they've ever made a demography-themed beer label? How could you not?
Happy 2015 everyone, and see you at PAA!
OK, I gave in. Not there is a Twitter account tracking Human Fertility Database updates, @HFDtracker
It does the exact same thing as @HMDtracker and will likely format in a similar way.
The HMD updates on a rolling basis, and until now the only update feed was the What's New page, or else by manually browsing the metadata on the website. Andrew Noymer and Mikko Myrsklä both requested an announcement service. While the HMD is pondering a rare newsletter to announce major changes, we can't exactly send out spam each time a series is revised. So, I set up a twitter account to automatically relay updates made to the What's New page.
The Tweets are dry and look like this:
So, just follow on Twitter to get updates
Details, in case anyone wants to replicate for other pages. It works like this:
1) track the What's New page using page2rss.com (spits changes into an RSS feed). You could just do this step and then subscribe to the RSS feed in your favorite blog reader...
2) pipe the RSS feed into @HMDtracker on Twitter using an account on twitterfeed.com
it's super easy and free to set up. I am not going to go and set up a system like this for each country in the HMD, but a motivated individual could certainly do so, and it would also work for any number of other oft-used data sources, such as the HFD, which also has a What's New page.
A few weeks ago I decided to send this (asocial, pseudo) blog's RSS feed to Twitter using step 2... works OK, not the prettiest tweets, but it's automatic anyway.
The other day Dr. James Carey gave a varied talk at Berkeley Demography. One of the awesome things he whipped out was this picture:
These are sketches of people, by people, from during the last ice age, ca 15000 BCE. Zoom in and guess how old they are? Some of those faces are of elderly people. In a CAVE in the ICE AGE. See more here. There were always old people, it's not some exclusively modern phenomenon, it's just that not so many people made it past childhood back then, so average lifespans were low.
I'm super excited to have just found out that I'll be able to present a paper at the New Measures of Age and Ageing conference in Vienna, December 3-5, 2014. Our proposal (this work is together with Pil H. Chung, Jeroen Spijker and John MacInnes) is tentatively titled: "Time-to-death patterns in markers of age and dependency". We're using the US Health and Retirement Study to document the thanatological age patterns (time to death patterns) of a large number of variables related to the measurement of ageing and dependency. It's pretty exploratory, and for that reason super exciting. If you're super duper curious about what we're up to, then all the goods are held in this repository on github. I'm pretty good about keeping things up to speed there, though the data aren't included because you need to register to use them.
The short proposal is attached at the bottom in case the following iframe doesn't render it properly in your browser:
Been working on some code to make ternary surface plots (or de Finetti surfaces?) made of tiled triangles. Here's a mock-up. The function plotted in this example is invented.
R code here. It's rather green (I notice the contours don't match expectations), and a work in progress, but I'm pleased so far
I really enjoy visiting poster sessions at conferences like the PAA or EPC. You get to meet lots of people and are exposed to tons of great ideas. I don't even bother checking the session topic or previewing contents-- I never regret going. One thing that annoys me is the unintentional swag you accumulate, specifically papers and shrunk-down posters that people give out. You can't read a poster on letter or a4 paper, so it's a waste of paper. So when I have posters I've tried just including links to related content in 2 ways, with varying degrees of "success" (meaning usage by visitors). I've settled on inserting QR-codes, like this:
But this is only useful if 1) there's wifi and 2) people have smartphones and are aware of this feature. The PAA lately has been very bad about wifi, FYI, but I suppose it depends on the venue... I've never seen a poster other than my own doing this, but I can attest that it was handy in several instances, and I'd recommend it.
Anyway, many people that came were curious about the QR codes and tried them. When wifi failed, some snapped pictures with their phones to capture the printed url (I use a url shortener to keep it simple). So, I recently got a question on how to put QR codes on posters and am going to give some simple steps here:
That's it! Paper saved! Dynamic content added to static paper poster! It'd be a good idea to have a tablet or similar on hand with the stuff preloaded. Maybe not a good idea to go overboard with this idea (put them off to a side?), but I think even if only one or two visitors make good use of the QR odes then it was worth the extra thought and effort. If you have a finished paper, consider linking to it as a pdf instead of printing it, or at least reduce the number of printed copies.