Chris Auld

Economics prof @UVic.
Econometrics, health.

[Chris Auld @Chris_Auld]

Breaking news! Using advanced multiple nonlinear regression models similar to those in recent news stories on alcohol and dairy and more than 3.6M observations from 1997 through 2012, I have found that drinking more causes people to turn into men!

[Chris Auld @Chris_Auld]

@iwelsh The top nominal marginal rate is not a measure of overall taxation conditional on income.

That said, your stats are also very misleading; marginal effective rates on very rich Canadians have been fairly flat since the 1940s, see e.g. Saez and Veall 2004:

[Chris Auld @Chris_Auld]

@PhilWMagness Steve Keen.

[Chris Auld @Chris_Auld]

@Sp0rtsFreq @ModestTeacher Suppose a billion people each shuffle 100 decks of cards a day every day for a thousand years.

Then the probability there were two identically shuffled decks is around 1 in 10 trillion trillion trillion (10^37).

[Chris Auld @Chris_Auld]

@Vince524 @CathyYoung63 That paper is famously flawed.

https://t.co/9Sm3q8mDZO

[Chris Auld @Chris_Auld]

The 2026 Canadian Health Economists' Study Group is May 27/28 in Victoria, B.C.

Ketnotes: Flavio Toxvaerd (@toxvaerd1) and Beate Sander (@BeateSander)

Program, registration, more info:
https://t.co/8awKRcwFZi

#HealthEconomics #EconTwitter #CanadianEconomics

[Chris Auld @Chris_Auld]

@_space_punk_ Narrowly self-interested players don't need to consider the actions of others: red is weakly dominant: you never do worse and might do better.

But, it's the same for perfectly altruistic players. Now blue is weakly dominant.

In between, you need to worry about others' actions.

[Chris Auld @Chris_Auld]

@HistoryBoomer @davidshor The game is called a "threshold public goods" game (contribute="press blue").

Lots of papers show higher stakes -> lower contributions.

We don't need to actually kill people to know that such enormous stakes would not yield the same outcome as zero-stakes Twitter polls.

[Chris Auld @Chris_Auld]

@RealDianeYap Agent i's preferences:

U(i)= D_i​+a_i​L_i​,

where Di=1 means i survives & L_i is the count of other survivors.

Higher a_i​ means "more moral" (altruistic).

The net gain from choosing blue is increasing in a_i​ whenever i might be pivotal.

ie, blue is "more moral," formally.

[Chris Auld @Chris_Auld]

@BarneyFlames There are many Nash equilibria in this game. One in which everyone on earth (N) pushes red, and many in which at least (N/2+1) push blue.

How do you know which of these many equilibria will be played, even assuming Nash play?

[Chris Auld @Chris_Auld]

@hecubian_devil 1. People are not narrowly self-interested.
2. This poll is terrible evidence in support of 1.
3. Communism doesn't work despite 1.

[Chris Auld @Chris_Auld]

@peterrhague A large experimental literature shows people are less likely to contribute in threshold public goods games as the stakes rise.

The red/blue button game is a threshold public goods game with extremely high stakes hypothetically and none on social media.

[Chris Auld @Chris_Auld]

@BarneyFlames Consider a population with altruism and irrationality. (Like the real one.)

Then:

Button A: nothing happens to you but you might cause billions of deaths

Button B: you might randomly die but you might save billions of lives

[Chris Auld @Chris_Auld]

@BobMurphyEcon It is altruism vs selfishness in the sense that narrowly self-interested agents would play red, but in a population with altruism sufficiently altruistic agents could rationally play blue. Consider the extreme case in which an agent _only_ cares about others' outcomes.

[Chris Auld @Chris_Auld]

@BobMurphyEcon Red is weakly dominant if you're self-interested, but it might be rational to play blue if you have altruistic preferences and believe there are a sufficient number of other people who will play blue for whatever reason.

[Chris Auld @Chris_Auld]

@cselley "This is awful, my shopping will be so slow."

[Chris Auld @Chris_Auld]

@HistoryBoomer The good news is many people holding such folk-economic beliefs actually respond to contrary expert information :

Can Economic Fact-Checking Remedy Incorrect Beliefs About Housing Markets? https://t.co/EXx0h8FGW6

[Chris Auld @Chris_Auld]

@RobLogic @grok I have failed to convince @grok this is not real. I enjoyed the bit where it told me I shouldn't believe Chat-GPT's diagnosis that it's fake because Chat-GPT is just an LLM.

[Chris Auld @Chris_Auld]

@JohnHolbein1 You wouldn't say that if you knew more math.

[Chris Auld @Chris_Auld]

@GLabsPlus @Math_files If Pr(boy)=0.5 & Pr(day of week)=1/7 & the mother always randomly selects a child, then

Pr(other child is a boy | reported child is Tuesday boy) = Pr(other child is a boy) = 0.5.

All other answers implicitly assume weird behavior on the mother's part.

[Chris Auld @Chris_Auld]

@flxia_ @FangYi11101 If we roll n die and sum the result, that sum has a distribution which looks more and more normal as n grows.

Here, OP drew 10,000 times from the distribution of the sum of n=2 die, but that sum does not follow a normal distribution.