Where communities thrive

• Join over 1.5M+ people
• Join over 100K+ communities
• Free without limits
Activity
alanlujan91
@alanlujan91
hi!
Mridul Seth
@MridulS
Hi!
Mateo Velásquez-Giraldo
@Mv77
$u'(c_t)=\beta R E[u'(c_{t+1})]$
Oh, I like this.
Can't figure out how to create new "conversations"
2 replies
alanlujan91
@alanlujan91
you can do latex!!? game changer
Mateo Velásquez-Giraldo
@Mv77
$y(e(s(\text{sir})))$
Mridul Seth
@MridulS
you want to get more crazy? This is federated via matrix so you don't even need to use a webpage to interact with gitter. You can use something like https://element.io
AMonninger
@AMonninger
Well, here is a question. Is there a HARK equivalent to the quantecon nelder_mead optimization function?
Unfortuantely, it only works with @nijt functions which I'm not familiar with.
4 replies
Mateo Velásquez-Giraldo
@Mv77
Hey, wanted to put this team on our radar https://github.com/SparseGridsForDynamicEcon . They are coming out with a continuous-time toolbox that might not be as relevant to us. However, the SparseGrid technology might be highly relevant, and they have examples on how to use it.
2 replies
Mateo Velásquez-Giraldo
@Mv77
Another tool I wanted to share: https://github.com/pytask-dev/pytask . It's a workflow tool aimed at facilitating reproducible analyses. It might be worth exploring it as a framework for the REMARKS that could replace the do_all/do_min setup
Durgesh Patel
@pat749
Hi, My Name Is Durgesh Patel and I am an Undergraduate Student Studying At NIT Trichy. Looking Forward to contribute on econ-ark . Thank you
Durgesh Patel
@pat749
Hi is there any progress on this (https://github.com/EconForge/heterogeneous-agent-computing/wiki ) documentation or should I work on it?
Sebastian Benthall
@sbenthall
I don't know what's up with that EconForge HACK group, but I don't think it's been active for some time. There is still a lot of heterogeneous agent modeling going on around here though.
Sebastian Benthall
@sbenthall
@pat749 if you are interested in helping on documentation, this issue needs some attention: econ-ark/HARK#1147
The issue is that there are modules that are in the HARK library but not yet in the API documentation: https://hark.readthedocs.io/en/latest/reference/index.html
That's because the documentation directory is missing the corresponding sphinx files: https://github.com/econ-ark/HARK/tree/master/Documentation/reference
Somebody needs to add .rst files to the Documentation/reference directories to get Sphinx to process the docstrings and produce autodocs
Durgesh Patel
@pat749
Ok so I will start working on this documentation if I need help I will talk through this group.
alanlujan91
@alanlujan91
wdu9
@wdu9
Hi, here is a link to a github repo that provides an easy to use toolkit to solve HANK models https://github.com/shade-econ/nber-workshop-2022
Mateo Velásquez-Giraldo
@Mv77
I've just merged a PR fixing the DemARKs that were broken by my Array-valued distributions: econ-ark/DemARK#185
AMonninger
@AMonninger

Hi,
I have a question regarding plotting solutions for different time periods. After solving a model, say IndShockConsumerType, how do I extract the consumption functions for different periods?
Take the example:

Import

import numpy as np
from copy import deepcopy
from HARK.ConsumptionSaving.ConsIndShockModel import IndShockConsumerType, ConsumerSolution, init_idiosyncratic_shocks
import time # to time

from HARK.utilities import plot_funcs_der, plot_funcs
import matplotlib.pyplot as plt

init_indshock = dict(
init_idiosyncratic_shocks,
**{

    ### No income shocks
"PermGroFac": [1.0],
"PermShkCount": 7,
"PermShkStd": [0.0],
"TranShkStd": [0.0],
"UnempPrb": 0.0,
"UnempPrbRet": 0.0,
"LivPrb": [1.0],
### For simulation
"aNrmInitMean": 0.0,
"aNrmInitStd": 0.0,
"T_sim": 5,
"AgentCount": 10,
### Others
"BoroCnstArt": 0,
}

)

Solve IndShockConsumerType

IndShockExample = IndShockConsumerType(**init_indshock)
IndShockExample.update_income_process()
IndShockExample.solve()

To get the consumption Function, I use:
IndShockExample.solution[0].cFunc

But how do I get the cFunc for different periods? Say t=0, 1, 2, 3, 4

Mateo Velásquez-Giraldo
@Mv77
That'd be in agent.solution[t]
@AMonninger the model you have above has default parameter IndShockExample.cycles = 1 which means it runs the solver once and thus only has two solutions, [0] is the solved solution and [1] is the terminal solution of consuming everything
If you go to https://github.com/econ-ark/HARK/blob/master/examples/ConsIndShockModel/IndShockConsumerType.ipynb and check out LifecycleExample you should be able to do what Mateo suggested