Lightnews — Scholar-powered news

James Bland

@jamesbland.bsky.social

950 followers 1.5K following 310 posts

Economist at UToledo. 🇦🇺 Bayesian Econometrics for economic experiments and Behavioral Economics Free online book on this stuff here: https://jamesblandecon.github.io/StructuralBayesianTechniques/section.html https://sites.google.com/site/jamesbland/ He/his

jamesblandecon.github.io

Posts Media Videos Starter Packs

Pinned

James Bland @jamesbland.bsky.social · Sep 14

On October 16th I will be presenting material from my book in the Workshops for Ukraine online workshop series. Details here 👇👇

sites.google.com/view/dariia-...

#EconSky #RStats #TeachBE

Dariia Mykhailyshyna - Workshops for Ukraine

Feedback on the past workshops (if you want to learn how to make wordclouds, check out Text Data Analysis workshop below)

sites.google.com

2 3 4

James Bland @jamesbland.bsky.social · 1d

An amazing paper, and related to some work I am doing on optimal experiment design. Well worth a read!

#EconSky

Alex Alekseev @aalexee.bsky.social · 5d

Thrilled to see my paper "The (Statistical) Power of Incentives" out at the Journal of the Economic Science Association. 🥳

Read it here (open access) 👉 dx.doi.org/10.1017/esa....

#Econsky

The (Statistical) Power of Incentives | Journal of the Economic Science Association | Cambridge Core

The (Statistical) Power of Incentives

dx.doi.org

1 1

James Bland @jamesbland.bsky.social · 1d

The take-away? For most participants we are probably OK in interpreting revisions as correcting mistakes. But this is not true for the entire sample.

James Bland @jamesbland.bsky.social · 1d

And about 75% of participants actually make themselves better off on average.

But that leaves about 25% of participants who made themselves *worse* off by revising their choices.

1 1

James Bland @jamesbland.bsky.social · 1d

I then use this estimated utility function to calculate the certainty equivalent of initial and revised choices.

For almost all participants, the revised choice is much more likely to be a utility improvement than not.

James Bland @jamesbland.bsky.social · 1d

I approach this problem differently: with structural estimation. Here, I make much stronger assumptions about the functional form of utility, but that lets me make stronger claims.

Specifically, I assume that participants have a rank-dependent utility function.

James Bland @jamesbland.bsky.social · 1d

Using indices that measure how consistent decisions are with maximizing a utility function, Breig & Feldman (2024) find that revised choices are generally more consistent than initial choices.

James Bland @jamesbland.bsky.social · 1d

In this experiment, participants made 50 convex budget choices that determined a risky payoff. Then, participants were given an opportunity to revise 36 of these.

James Bland @jamesbland.bsky.social · 1d

I use structural techniques to estimate the utility gain (or loss) associated with these revisions in an existing economic experiment (Breig & Feldman, 2024).

James Bland @jamesbland.bsky.social · 1d

Sometimes in experiments, we give participants an opportunity to revise their choices. These revisions are often interpreted as revisions of mistakes.

James Bland @jamesbland.bsky.social · 1d

New working paper: The normative value of a revised choice: a
structural approach

papers.ssrn.com/sol3/papers....

#EconSky

A thread ...

1 5

James Bland @jamesbland.bsky.social · 3d

Thanks to the @ecscienceassoc.bsky.social team for an amazing N American meeting. I am glad that there is room for me in this space!

Reposted by James Bland

Andrew Gelman et al. @statmodeling.bsky.social · 4d

7 reasons to use Bayesian inference!
statmodeling.stat.columbia.edu/2025/10/11/7...

7 reasons to use Bayesian inference! | Statistical Modeling, Causal Inference, and Social Science

statmodeling.stat.columbia.edu

1 11 23

Reposted by James Bland

Dariia Mykhailyshyna @dariia.bsky.social · 5d

❗️Our next workshop will be on Oct 16th, 6 pm CEST titled Structural Bayesian Techniques for Experimental and Behavioral Economics in R& Stan by @jamesbland.bsky.social
Register or sponsor a student by donating to support Ukraine!
Details: bit.ly/3wBeY4S
Please share!
#AcademicSky #EconSky #RStats

1 13 12

James Bland @jamesbland.bsky.social · 6d

One week until this online workshop on my book!

James Bland @jamesbland.bsky.social · Sep 14

On October 16th I will be presenting material from my book in the Workshops for Ukraine online workshop series. Details here 👇👇

sites.google.com/view/dariia-...

#EconSky #RStats #TeachBE

Dariia Mykhailyshyna - Workshops for Ukraine

Feedback on the past workshops (if you want to learn how to make wordclouds, check out Text Data Analysis workshop below)

sites.google.com

James Bland @jamesbland.bsky.social · 10d

Proposed new language for "which is how it is done in the literature"

a cartoon of homer simpson with the words which was the style at the time

ALT: a cartoon of homer simpson with the words which was the style at the time

media.tenor.com

James Bland @jamesbland.bsky.social · 17d

Is this something you could feasibly do? Yes! My computer designed these experiments in less than a day. It is cheap compared to paying for a less good experiment.

James Bland @jamesbland.bsky.social · 17d

So should I use this technique for designing experiments? I would say yes with a but: Yes, but the algorithm only understands the econometrics. It doesn't really understand humans the same way experimenters do. So in the end it should just be used as a suggestion.

James Bland @jamesbland.bsky.social · 17d

Admittedly I don't gain any intuition along the way, but my computer provides suggestions for designs that I couldn't have thought of myself.

James Bland @jamesbland.bsky.social · 17d

The take-away? The best experiment design depends on the research question. In some sense: duh. But in another sense going into this I had absolutely no intuition as to how these should differ.

James Bland @jamesbland.bsky.social · 17d

The designs are really different to each other, and also look (to me at least) very different to existing designs by humans.

James Bland @jamesbland.bsky.social · 17d

Each of these experiments consists of 80 pairwise choices over lotteries with 3 prizes. There are 320 design variables!

James Bland @jamesbland.bsky.social · 17d

- One to estimate all of the parameters as precisely as possible,

- another to estimate just the probability-weighting parameters as precisely as possible, and

- another to estimate a certainty equivalent.

James Bland @jamesbland.bsky.social · 17d

I demonstrate this approach by designing three experiments to estimate a rank-dependent utility model. ...

James Bland @jamesbland.bsky.social · 17d

Once we've written down (and probably also approximated) our utility function, we then need to maximize it. This is computationally non-trivial: the choice set for experimenters is very large! I tackle this with an exchange algorithm.

James Bland @jamesbland.bsky.social · 17d

Furthermore, in structural estimation we are often interested in transformations of the parameters. What if we wanted to estimate one of these transformations well? My approach also lets you do that!