search

QuantEcon / QuantEcon.notebooks

Jupyter notebooks contributed by QuantEcon developers, users and the community
BSD 3-Clause "New" or "Revised" License
122 stars 105 forks source link

Add python/julia codes for solving NK Model in CLMM #53

Closed sglyon closed 7 years ago

sglyon commented 7 years ago

In this PR I included two notebooks that serve as part of the appendix for a working paper I'm working on with @cc7768 and @maliars.

The notebooks provide code for solving a New Keynesian model with an option to impose or not impose a zero lower bound on interest rates.

Once the paper has a home somewhere online I will update these notebooks to include a link to the paper.

@cc7768 might have a couple more notebooks coming that are the other part of our Python/Julia appendix demonstrating 7 different algorithms for solving a standard growth model.

thomassargent30 commented 7 years ago

This would be a very good addition

On Jun 3, 2017 12:03, "Spencer Lyon" notifications@github.com wrote:

In this PR I included two notebooks that serve as part of the appendix for a working paper I'm working on with @cc7768 https://github.com/cc7768 and @maliars https://github.com/maliars.

The notebooks provide code for solving a New Keynesian model with an option to impose or not impose a zero lower bound on interest rates.

Once the paper has a home somewhere online I will update these notebooks to include a link to the paper.

@cc7768 https://github.com/cc7768 might have a couple more notebooks coming that are the other part of our Python/Julia appendix demonstrating 7 different algorithms for solving a standard growth model.

You can view, comment on, or merge this pull request online at:

https://github.com/QuantEcon/QuantEcon.notebooks/pull/53 Commit Summary

  • Add python/julia codes for solving NK Model in CLMM

File Changes

Patch Links:

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/QuantEcon/QuantEcon.notebooks/pull/53, or mute the thread https://github.com/notifications/unsubscribe-auth/AD22Tu1FM5IDZtse29gNpkt_JG1Wj26Vks5sANrvgaJpZM4Nu91u .

cc7768 commented 7 years ago

These notebooks are excellent. There is lots of useful code, but seem a little sparse in terms of words. If these will be "stand alone" notebooks on the QE notebooks site, it might be helpful to provide a little more description of the model and solution. (I guess you could counter this suggestion by saying that the details are in the paper and can be found there).

sglyon commented 7 years ago

I was also a little torn about having np description of the model, but after having thought about it I see it as more of an advantage. This is an online appendix and the paper contains the model.

If you think it would be better I would be happy to include the list of equilibrium conditions and variables (similar to what we do in the main body of the paper right now). Thoughts?

jstac commented 7 years ago

@sglyon Many thanks. Very interesting.

In my view we could do without all the equations as long as we have a link to the paper embedded in the notebook. Do you have a permanent URL?

Also, at the end of the computations I was looking forward to seeing interpretation. I suppose that's in the paper and if so that's fine --- there's no need to repeat. But perhaps at the end of each notebook a line that says interpretation can be found in the paper would be good.

sglyon commented 7 years ago

Hey @jstac that's a good idea. I'll at least add a line directing people to the paper if they want to learn more and see some interpretation.

Once we hear @cc7768's vote about the equations, we can make final decision about that.

We don't yet have a permanent URL, but when we do I'll revise these notebooks and put it in.

cc7768 commented 7 years ago

I could live without the equations if we eventually can point to a place where people could investigate if they were interested in them.

I guess my opinion that it is sparse on words depends on whether this is meant solely as a computational appendix (i.e. a place for people to read and explore the code after having viewed the paper) or whether this is meant to be a standalone notebook people may stumble across on QE. My interpretation was that we would hope for interest generated in both directions, if that is true, I think the notebook could use a few more words. For example, in the "Solution Routine" section there are no words that describe what is happening. A couple words in the solution routine section and a "conclusion" paragraph would add some value (even though a lot of it would already be said in the paper).