Andy Seaton

Its times when I let my actual sense of humour show that I consider deleting all social media. Considering that right now

Just as I must keep the women busy with semen, so shall I also make sure that I write that paper I've been procrastinating on.

Reading some Montaigne and mostly I enjoy his little melancholic meanderings. Then sometimes he casually drops in shit like this.

In the short essay On Idleness

Optimism*

Your level of optimising perfectly matches my own when I do statistics so I appreciate that.

Would be interesting if we could come up with something "robust to availability misspecification". No idea what that would look like though

Thanks Theo I'll check those out. The thing I never wrap my head around is the identifiability of each process. I've seen people use average moving speed to define availability, which makes some sense to me. Only using locations of animals seems totally confounded with selection though

I sometimes see this defended as 'we can't make progress unless we agree on the rules of the game'. And, as you say, if you don't agree with the rules, it's not seen as 'playing the game', it's seen as 'trying to ruin the game for others'.

Dunno if I'm being patronising putting it like that...

Have idly pondered if people have tried joint likelihood approaches like:

L(resource selection | availability)*L(availability)

But easier said than done I'd imagine.

And resource selection and availability are often confounded in real-world applications.

I'd like to see people being more open about analyses being conditional on *one specific choice* of availability.

You are saying "people feeling the squeeze and having to change spending habits" has no affect on the narrative about inflation?

Nothing radicalised me more than realising that part-time nursery costs for one child the UK cost more than our mortgage.

Thankfully we are out of that period now but for the best part of a decade getting part-time childcare was equivalent in cost a second-mortgage on a bigger house than our own

How does it look when inflation is compared vs wage growth?
It's more about if income keeps up with prices. And people know if they are cutting back, they know they aren't going to restaurants as much, or going on cheaper holidays, or the have to be more careful when buying groceries.

Bayesians who think like this must be exhausted. So much philosophical weight on their shoulders. Maybe we need to start running some wellness retreats. Yoga followed by rhythmic chanting "I do not need to solve philosophy using maths"

It's going to be kind of ironic when "some people are just too greedy for it to work" is used to describe capitalism instead of it being the go to folk criticism of communism...

Book cover for https://oliviergimenez.github.io/banana-book/

📘 New book out soon !

I’m excited to share that 𝐁𝐚𝐲𝐞𝐬𝐢𝐚𝐧 𝐀𝐧𝐚𝐥𝐲𝐬𝐢𝐬 𝐨𝐟 𝐂𝐚𝐩𝐭𝐮𝐫𝐞-𝐑𝐞𝐜𝐚𝐩𝐭𝐮𝐫𝐞 𝐃𝐚𝐭𝐚 𝐰𝐢𝐭𝐡 𝐇𝐢𝐝𝐝𝐞𝐧 𝐌𝐚𝐫𝐤𝐨𝐯 𝐌𝐨𝐝𝐞𝐥𝐬: 𝐓𝐡𝐞𝐨𝐫𝐲 𝐚𝐧𝐝 𝐂𝐚𝐬𝐞 𝐒𝐭𝐮𝐝𝐢𝐞𝐬 𝐢𝐧 𝐑 𝐚𝐧𝐝 𝐍𝐈𝐌𝐁𝐋𝐄 is being published by Chapman & Hall / CRC Press

Hope it’s useful to students, researchers, and practitioners

#StatisticalEcology #NIMBLE

Sometimes there's just no easy way to turn a big number into something intuitive. But I always enjoy when people attempt it, no matter how successful they are.

That's a good example. But have you considered communicating the total energy consumption of Scotland per year (approx 135 TWh) as

"roughly equivalent to boiling a household kettle 720 billion times"

(We joked about doing this when I worked in Energy Stats for Scot Gov)

"the Falkirk Wheel uses only 1.5 kWh of energy to turn, the amount it would take to boil your household kettle 8 times"

The Falkirk wheel lifts canal boats 24m from the Clyde canal to join with the Union canal Scotland. 8 kettles to lift a boat along with a bit of attached canal is awesome.

Quote from this:
www.theguardian.com/news/ng-inte...

My personal experience of 3 million Olympic-sized swimming pools really helps me here.

I'm being sassy here but I'm not joking that I like to collect these examples. A good example is below:

I like to collect examples of people trying to make numbers feel intuitive by relating them to things in everyday lived experience. For example:

"The extra water falling in the UK each year was equivalent to 3m Olympic-sized swimming pools, Fowler said: 'That’s a lot of extra water'"

Reproducing kernel methods for machine learning, PDEs, and statistics This monograph develops a unified, application-driven framework for kernel methods grounded in reproducing kernel Hilbert spaces (RKHS) and optimal transport (OT). Part I lays the theoretical and nume...

Have you seen this beast: arxiv.org/abs/2402.07084

I read it a bit a few years ago but have never gone through it in detail. Just a brief 180 pages...

Got it, I think I don't understand where a Jacobian would come into it but I should probably just go and read the docs.

I am fond of this perspective since it was my way into the literature in the first place. I started with INLA, then the SPDE approach, then Dave Miller convinced me to try and recreate it in mgcv. It was a very informative exercise for me.

But I'm not sure I've seen any systematic discussion of situations like this.

I think I've seen stuff like

1. Don't define the mesh where the animal wont go (not a great idea if you are wanting to infer species-habitat associations...)

2. Include a factor covariate and hope the estimate is strongly significant so an apparent hard threshold appears in the predicted intensity

Also thanks for taking the time to reply. It's fun being online again and talking stats. I've been out of the game for 3+ years for health reasons, so I appreciate it a lot!

Got it, will keep this in mind. Hard to know how much I will be allowed to harp on about this stuff in our paper but I would like to get something in there if possible.

Are people in fisheries doing spline stuff in constrained parameter domains where they need a transformation to something that plays well with stan? Don't know much about stan or fisheries.

I also have strong opinions! Then I look at what mgcv does and what, e.g., INLA does and I go ach it's not so different really. Thinking about it properly feels really important, but it in many cases it may not lead to anything consequentially different in terms of inference...

Thanks, I will have a read. The title alone it reads like something I want to say is not quite true! But it's so useful to act like it's true when learning about the methods, especially if someone is already familiar with GPs or splines.