cycles = 0

[alanlujan91]

No description provided.

[alanlujan91]

Using dask to speed up numerical exercise; parallel processing of function calls scaled up to computer resources.

[sbenthall]

cool.

let me know when this is ready to merge.

[alanlujan91]

@sbenthall i removed some of your code, especially the plotting, to make this faster. Is that ok? Do we need some sort of plot for this?

[sbenthall]

i think the plotting was just an artifact; no worries

[sbenthall]

Is this supposed to be the aNrmInitMean ?

[alanlujan91]

I didn't add this here, but I think it should be setting aNrmInitMean

[sbenthall]

I agree. Would you mind correcting that? I would do it myself but it's in your personal PR.

[sbenthall]

This value is 6 in the Numerical notebook. Which do you mean?

[alanlujan91]

it is 5 in the default so let's go with 5

[sbenthall]

is this ready to merge?

[alanlujan91]

Need to actually calculate aNrmInitMean, which notebook does this?

[sbenthall]

I don't think we've settled on a process for this.
There is code that does something like this in the numerical buffer stock and dashboard notebooks.
But there's nothing that says specifically:

• here's the parameters we are committed to
• here's the calculations we are doing to flesh out those parameters.

certainly open to improvements.

[alanlujan91]

Ok ready to review

[sbenthall]

Just to double check:
... are these dividend growth and stdev rates correct? i.e., did you use the same ones when calculating the target wealth?

https://github.com/sbenthall/SHARKFin/blob/master/simulate/run_any_simulation.py#L109-L118

honestly it might be better to have all these details settled in one document.

the configuration is scattered across many different points but we need a canonical description of what we're trying to do. This info will also go into the paper.

[alanlujan91]

@sbenthall on scaling... we can't scale income.... mNrm_next = a * R + 1*p_shk, we would need mNrm_next = a * R_scaled + 1/4 * p_shk_scaled

[sbenthall]

Could we:

• start with a pLvl scaled down?
• scale the income shocks down appropriately?

Not sure I'm following the logic here, but maybe you can run the question by Chris?

[alanlujan91]

@sbenthall ready to review

see macro/macro_parameters.py for new master list of parameters, annual and quarterly

[sbenthall]

Ok, merging now, thanks.