Daniel Lakeland

https://mastodon.sdf.org/@dlakelan

Applied Mathematician, Julia programmer, father of two amazing boys, official coonhound mix mutt-walker.

PhD in Civil Engineering. Debian Linux user since ca. 1994.

Bayesian data analysis iconoclast

Github: https://github.com/dlakelan
OpenWrt: https://forum.openwrt.org/u/dlakelan/preferences/profile

Public Key

npub1fnejmanw24sxweer9t5ncla5v5ekdf6mvtwnjmkm7em6vxjnxf6swp87ll

NIP-05 Address

dlakelan_at_mastodon.sdf.org@momostr.pink

Profile Code

nprofile1qqsyeued7eh92cr8vu3j46fu076x2vmx5adk9hfedmdlvaaxrffnyagmqugsh

Author Public Key

npub1fnejmanw24sxweer9t5ncla5v5ekdf6mvtwnjmkm7em6vxjnxf6swp87ll

Show more details

Last Notes

2024-06-16 21:50:24

"The American Economy is doing so well!!!"
Alternatively "the US economy is a shitshow and has been for... (checks graph) 50 years."
https://mastodon.sdf.org/system/media_attachments/files/112/628/487/985/479/354/original/ed03ffc8d7334001.png

2024-06-06 14:37:26

- reply

I just don't see a kinetic dance floor as likely to capture much energy, enough to counteract the cost of disassembly, transport, and reassembly seems unlikely. Energy is force times distance. Dance floors likely support say 100lb per sqft, assume a sine wave of 1 inch displacement at 1hz, let's call it 100 inch pounds per second, and a 4 hr concert per day and let's say 10% capture... A 1000sqft dance floor would put out about 200 watts, or power 10 phone chargers

2024-06-06 14:17:39

- reply

Isn't 105% of this reduction the train? (The dancefloor and LEDs are probably a net increase in carbon)

2024-05-28 20:55:01

- reply

I would argue we always want dimensionless ratios and usually we want time averages over a fraction of a lifetime. A reasonable one to care about is smooth(income) / smooth(expenditure rate) for expenditure rate being the cost of things like housing food transportation, health, and childcare and the smoothing function being maybe an exponential moving average with a timescale of say 0.05 lifetimes which is more or less 4-5 years.
@npub165c…tdlq

2024-05-28 20:23:45

- reply

Just say "rate".
Consider the following two sentences.
"Inflation has eaten away at my ability to afford childcare"
"Inflation rate has eaten away at my ability to afford childcare"
I think the first sentence is perfectly cromulant and doesn't mean the same thing as the second which is actually nonsensical.
This suggests that "inflation" doesn't mean "inflation rate" even to economists. Right?
@npub13xd…hkyz

2024-05-28 15:51:46

- reply

The thing that concerns me most is that it looks, based on the trend line, like if prices held completely constant, it'd take about 10-12 years to have prices catch up to the trend line... That's a bad situation, and the people talking about how the economy is great in the media should shut up.

2024-05-28 13:36:35

- reply

I think a lot could be explained by the concept of excess inflation, similar to excess deaths. Everyone has some notion that prices slowly go up but when they go up faster than expected then the gap between where people feel the price level should be and where it is is considered excess inflation. I drew the highlighter line with my thumb on my phone so a little charity is needed for the fit 😅
https://mastodon.sdf.org/system/media_attachments/files/112/518/980/885/446/056/original/0259eeb6406f066e.png

2024-05-24 00:07:53

- reply

Same could be said for any 10 who talk about how great it is. It's actually called out in the merriam-websters dictionary entry as one of the most frequently looked up words because it's so abstract and has so many different meanings to different people.
https://www.merriam-webster.com/dictionary/capitalism

2024-05-23 22:47:41

- reply

To me this is the big question when "capitalism" comes up a lot. IMHO any 10 people in a room are likely to have at least 4-5 varying meanings for this word. Some just think it means "market economy" but I reject that. It must mean something more specific than that. I'm really not sure what most people think though.
@npub182r…pm8s @npub165c…tdlq

2024-05-23 22:42:56

- reply

You are using the word "imply" as if it were the word "suggest". Yes, correlation suggests some causation *may* be at work, but the word "imply" in "correlation does not imply causation" means "logically prove". As in big feet implies small shoes will not fit.
@npub182r…pm8s @npub1gg4…euuk

2024-05-23 22:39:50

- reply

It's easily possible to make a system controlled by a variable whose value is completely uncorrelated with the variable it controls...
Drive over hills while using cruise control. The throttle pedal controls the speed and yet while the throttle goes up and down, the speed remains constant. There is zero correlation between cause and effect.
@npub182r…pm8s @npub1gg4…euuk

2024-05-22 23:15:01

I give you civilian labor force with a disability (16yo or older)
https://fred.stlouisfed.org/series/LNU01074597
https://mastodon.sdf.org/system/media_attachments/files/112/487/274/631/964/922/original/1a88df73b7d613ff.png

2024-05-21 16:17:05

- reply

There's a lot of talk about capitalism this, and socialism that. But most people hear "businesses owned by non-govt people" vs "businesses run by an authoritarian govt like in USSR".
Those are the two options people think we have, and so understandably they're not in favor of the second one... If we could explain better to everyday people we would get farther. Not only are those not the only options we don't even have either one right now.

2024-05-21 12:59:49

- reply

That's exactly why I'm not holding my breath. You can see how much The Limits To Growth was revered as effective and realistic research /s
@npub165c…tdlq

2024-05-21 06:15:42

- reply

Now you can see why I keep thinking I need to start a nonprofit and get funding for research because this kind of question is the kind of question I want to ask and start to answer... Know anyone looking to fund nonprofit research into these questions?
@npub165c…tdlq

2024-05-21 06:01:38

- reply

This is ultimately intimately connected to civil engineering. If we stop chasing profit and start doing nonprofit stuff or worker owned coop stuff based on say anarchic councils on investment, then we will need answers to questions like "given that people have voted to build a combined living space and vertical indoor farm, how can we best design and build it to maximize well being return on usage of physical resources?" In other words civil engineering.
@npub1gg4…euuk

2024-05-20 13:14:32

- reply

I don't really know what goes into business degrees. But the only person I know who has one runs a small business providing gymnastics instruction to kids ages 2-18. So let's do more of that please.
@npub12p6…gpnr

2024-05-15 20:26:01

I haven't been following daily polls or anything like that, but if you were to ask me Trump vs Biden, I'd say Biden vote fraction of the two party vote approximately ~ Normal(50,3) (ie. 50% +- 3 percentage points standard deviation)
If you look at highly sophisticated The Economist poll modeling... It's basically ~Normal(50,3) so polls offer essentially zero information above and beyond "hey they're about equal" which we knew before looking at any information at all. #statistics #voting #poll
https://mastodon.sdf.org/system/media_attachments/files/112/446/973/226/103/542/original/f80279c4f297ae8b.png

2024-05-06 21:13:43

- reply

You had me going there for a moment...

2024-05-06 14:03:07

- reply

What does a more viable logic look like?
Having hypothesized a mechanism for how MY DATA comes about, and explained that in terms of a simplified model with CERTAIN PARAMETERS and collected data from these processes, does the mechanism and the data imply different parameters for my different experimental conditions? Given the results of those analyses, do the models consistently correctly predict future data?
This is the logic of quantitative model building and #Bayesian statistics.

2024-05-06 14:03:05

- reply

So the very basic stuff you are taught in masters level #biostats or #socialsciences stats or whatever is built on two layers of logical fallacy.
So that's why poor stats practices are absolutely more common than good stats practices. That's why the kind of thing posted on the Reddit post I mentioned in my recent posts is actually everywhere in science. Students are literally taught to do things wrong in the textbooks.

2024-05-06 14:03:03

- reply

The very first basic logical assumption is invalidated... But even if we grant that's true and that you go around studying random number generators, a class of things so notoriously hard to build that they are repeatedly the cause of major security flaws in cryptographic software between 1990-2010, the existence of a small p value in a test says NOTHING about the explanation for WHY the test fails, it particularly says nothing about whether your favorite explanation is true.

2024-05-06 14:03:02

Here's the logical structure of what you will be taught in terms of #statistics as a masters student in pretty much any #science field.
If MY DATA is a sample from two random number generators of PARTICULAR TYPE, and MY TEST has a small p value then MY FAVORITE EXPLANATION FOR THE DIFFERENCES IS TRUE.
This is, quite simply, a logical fallacy. The first thing wrong is that your data IS NOT a sample from a random number generator of that particular type. So we can ignore the rest logically.

2024-05-06 03:17:40

Here it is people. A PhD student describing details of what they've come to realize is the completely scientifically bankrupt methodologies their high-powered successful, well funded lab PI demands the lab members do. Everything this person says is basically commonplace in todays labs #science #openscience #statistics #bayesian
https://www.reddit.com/r/PhD/comments/1cksfmd/i_realised_that_my_pi_and_research_group/

2024-05-03 15:53:17

- reply

Also, If you're not following Kyle, you should be. We need economists who think outside the current box.
@npub1mxw…n7uy @npub1v0m…7sw2 @npub1wsc…89rx @npub1zwx…k383 @npub1thv…jkz4 @npub13xd…hkyz

2024-05-02 22:48:43

- reply

What the rescheduling does is it lets you go to your Primary Care Physician in a legalized state and say something like "using marijuana is effective against my anxiety/depression/nausea/whatever can you prescribe me some?" and the doctor can legally prescribe it for you in the same way that they can prescribe codeine after tooth extraction. And then it's not a federal law violation to have it so you are 100% legal.

2024-05-02 16:14:32

- reply

Oh Shit, how could I forget sports? Large universities are Research first, sports second, dorms and amenities third, providing tax dodges to the ultra wealthy fourth, and teaching a definite fifth.

2024-05-02 14:56:50

- reply

What's the value that UCLA or USC offers over CSU LA or CSU SF? It's *the selectivity*, the elitism, the fact that you can only take their courses if you "got in" and your cohort has been rubber stamped as "high achieving" and such. It's a sickening bastardization of an educational mission. Signaling rather than education.

2024-05-02 14:44:08

- reply

The vertical consolidation is in the *degree* and its requirements.... you must take a bunch of specific classes that the university has prepared as a batch in order to get that degree. The degree serves no purpose other than to capture the student into the universities pipeline of money extraction. Seriously, if you took 20 or whatever courses from 8 different facilities do you learn less than a "degree" from a university?

2024-05-02 14:42:34

- reply

Large universities are research first, dorms and amenities second, teaching third or fourth. Students can get together at a teaching specific community college organization (a few tens of buildings) just as easily as the few tens of teaching buildings on large university campuses. In fact, they could go to different classes at different local locations if it weren't that the whole thing is "vertically consolidated".

2024-05-02 14:40:14

- reply

My point is against the *consolidated* large university, not people meeting in person.

2024-05-02 14:35:27

- reply

Certainly 30 years ago, running colleges had a reasonable argument to be made for it. But at this point in the evolution of technology I'm not even sure that running large Universities **at all** is a good idea. I think med-schools are a bit special, but as actual educational mission maybe vastly expanding the number of "city colleges" and allowing people to use their buildings and things for independent seminars etc would be better. Lectures on Peertube...

2024-05-02 13:00:19

- reply

Tuition costs inflated dramatically faster than any other area of the economy since 2000, and administration is like 3x as big as it was per student when I was in college the first time in 1990s. Meanwhile teaching has shifted to adjunct faculty who make less per hour than wait staff at restaurants. People outside of academia just imagine universities are similar to the way they were 30 years ago but its not even close.

2024-05-02 12:53:13

- reply

Its not the investments in Israel they're worried about, its losing donations from major pro-israeli donors I guess? Universities are disgusting. There's nothing good about them. I'm an anti-institutionalist. Education through decentralized collaboration, not power concentrating money sucking sheepskin granting rubber stamping organizations. I have never ever displayed a degree certificate granted to me by any of the 4 institutions I've attended.

2024-05-02 12:44:48

Seriously how profitable are investments in Israel that universities don't want to divest badly enough that they are willing to ghoulishly bring in police to fire baton rounds at peaceful 20 year olds? They will be on the wrong side of history and be eventually immortalized on the walls of the university halls as the people who fucked up... But they'll be responsible for 3% increase of the endowment!
I can't even...

2024-05-01 17:08:52

- reply

Is he on Mastodon? Love a lot of his videos.

2024-04-30 21:02:27

- reply

That's been my rant for the day and now I'm tagging [@economics-that-works](https://a.gup.pe/u/economics-that-works) at the tail end of this rant... If you're interested in the question of funding public goods via "printing" money and whether we should have banks involved in money creation at all, and student loan forgiveness and all that, you'll want to click and read from the beginning. thanks for coming to my TED talk.

2024-04-30 21:00:32

- reply

No bank should be saying "well, how much do you imagine you can charge for your public YouTube video on traveling flames Mr Steve Mould? (https://youtu.be/SqhXQUzVMlQ?si=IvbHaw-SSZ3-cYsq) The answer is, fuck charging for that. And fuck "advertiser supported", just **PAY STEVE TO MAKE THIS GREAT CONTENT AND LET EVERYONE WATCH IT**. And get Google out of the middle of it. And simply "print" the money to provide the grant to get him to do it (ie. manufacture it out of thin air) the same way we "printed" the money for GWOT.

2024-04-30 20:57:33

- reply

A vast fraction of the best things we can produce in society are *public goods*. Ignoring the bullshit we have developed in terms of copyright and patent law... Think of movies, books, music, the linux kernel, Arduinos, Julialang, Python, Jupyter, Geogebra, public education, public concerts, YouTube videos about science and history and economics... These are all non-rivalrous... we can all consume them together once created. So we should be *funding them* by *creating money* as *grants*.

2024-04-30 20:54:01

- reply

Commodity money (ie. gold or silver or whatever) needed banks because you needed a fucking fortress to actually keep people from running away with the gold bars. But we don't need that anymore with fiat money. Fiat money is *money* not *something as money* it's just serving the money function, and the bank vault is largely pointless, the function that banks serve is pretty much *third party accountants*. They're also serving this outdated function of "deciding who to loan to".

2024-04-30 20:51:02

- reply

When you create money and give it to rich people, rich people buy up capital assets with it, consolidating businesses, stuff like "private equity firms". These seek out short term financial gains rather than long term *real asset* gains.
Here's mortgage asset level per capita CPI adjusted... To get mortgage loans you have to be wealthy... So again wealthy people buy up houses, which we aren't constructing much anymore either.
https://mastodon.sdf.org/system/media_attachments/files/112/362/137/123/240/534/original/e34eaf3b213e1c20.png

2024-04-30 20:46:09

- reply

A huge huge huge part of why we have such a non-functional economy with wealth inequality so high is because *banks create money* but they give it almost exclusively to rich people. In this graph you can see commercial debt, CPI adjusted, per capita, through time. Repeatedly we give brand newly created money (aka Loans) to business owners, they fuck shit up, then we bail them out again, it's an amplifying oscillation. https://fred.stlouisfed.org/graph/?g=1lz5V
https://mastodon.sdf.org/system/media_attachments/files/112/362/032/571/151/819/original/048e43f0746fc874.png

2024-04-30 20:16:06

I'm against student loan forgiveness, not because I'm against giving the money out... but because I'm against giving it out *only* to student loan debtors. Here's the idea... give everyone equally a #UBI and let the people with student loans repay it from there. I'm ok with limiting interest rates too, and rolling back capitalization of back interest... totally fine with that. But if we're creating money out of thin air and giving it to anyone, it should 100% be equally distributed.

2024-04-23 19:39:27

- reply

keen seems really great in this short Youtube interview https://www.youtube.com/watch?v=1y3NSx8wYZQ so yeah, I"ll try to find out more about him. I'm so done with the equilibrium non-dynamic schools of thought, it drives me up the wall.

2024-04-23 18:28:23

- reply

Who are the heterodox economists you follow? Either on Mastodon or more generally in terms of reading their publications or watching their videos or etc?

2024-04-23 18:21:16

- reply

It feels to me like Economics was actually born in 1972 when The Limits To Growth was published. For the first time here was a general dynamical systems view of interacting stocks and flows and soforth... The Economics profession seems to have assiduously refused to notice.

2024-04-23 18:16:39

- reply

[@GhostOnTheHalfShell](https://masto.ai/@GhostOnTheHalfShell)
Just started skimming it. Jesus Christ it's worse than I thought. Economists are like at the stage where they've discovered mass... Look, if you pour water into a cup it doesn't matter what shape the cup is it weighs the same if it has the same volume... That people were still fundamentally sure that the concept of money was tied to gold as recently as the 60s is wild!
[@economics-that-works](https://a.gup.pe/u/economics-that-works)

2024-04-23 17:51:56

- reply

What's your take on the linked book? Or has the rest of your thread just not propagated to my server yet? 🙂

2024-04-17 15:49:23

- reply

[@FantasticalEconomics](https://geekdom.social/@FantasticalEconomics)
Agreed, I think willingness to pay is caused by value. But it's also caused by asymmetry of information, and scamminess. IMHO the best measure of value would be willingness to pay in a transparent world where all relevant facts are known by all parties. It's unobservable, but it's what we mean by value. As a Bayesian an unobservable doesn't bother me much.
[@jackofalltrades](https://mas.to/@jackofalltrades) [@droidboy](https://social.cologne/@droidboy)

2024-04-16 20:04:57

The Ratio of CPI all-items to the equal weighted average of 5 items: rent, food, transport, childcare, energy.
When this goes down it means the 5 item index grows faster than the "overall average index" (ie. "families real needs becoming more expensive faster than the CPI suggests"). When it goes up it means that other expenses outside of family core needs go up faster.
https://mastodon.sdf.org/system/media_attachments/files/112/282/649/352/469/721/original/ee30237040c34f2c.png

2022-12-11 18:33:34

#introduction when the database crashed I lost some followers, but because I hadn't backed up my account yet, I am not sure who they all are. As an extended introduction I'll tag some common topics I like to discuss, maybe people will see those.
#bayes #bayesian #statistics #appliedmathematics #ipv6 #bufferbloat #quarto #openscience #agentbasedmodels #julialang #economics #biology #molecularbiology #ecology #forensics #engineering