I have had some interesting conversations in the last couple weeks about this question: Should we want the equilibrium of our trade model to be unique? raw.githack.com/jdingel/econ... - ThreadSky

About ThreadSky

tradediversion.bsky.social • 547 days ago

I have had some interesting conversations in the last couple weeks about this question: Should we want the equilibrium of our trade model to be unique?

https://raw.githack.com/jdingel/econ6905/main/slides/week01/slides.pdf

Comments

veryshuai.bsky.social•545 days ago

Structural models allow one to (1) measure the strength of forces in your model, and (2) run counterfactuals. If there is multiplicity, (2) is more difficult. We are just starting a project and trying to decide whether to use a richer model with mult eq. or a simpler model w/o.

veryshuai.bsky.social•545 days ago

Gets close to the hard nut: "What is the goal of inquiry?". Haile says that structural modeling is all about cfs. Love your slide deck btw, will steal (with credit) part for some teaching I'm doing this semester.

jlperla.bsky.social•545 days ago

Hysterisis and multiplicity are rampant in the world, we can't just assume it away by choosing models where it isn't possible. It has its place, we need to make sure we don't choose uniqueness for the wrong reasons (ie, numerical or analytic tractability).

aloneizenberg.bsky.social•545 days ago

The IOish stance has been that we can learn a lot from necessary equil conditions which hold under multiplicity. For counterfactuals there are learning algorithms that pick a unique outcome. Is this complexity justified? Yes if allows us to ask more questions or relax untenable assumptions.

Posting Rules

Be respectful to others
No spam or self-promotion
Stay on topic
Follow Bluesky's terms of service

Comments

Posting Rules

Reply