Comments (9)
This applies to:
- density
- plot (parameters, irfs and fcasts)
- as.mcmc
- prep_data
- get_var_set
In general we have to differentiate:
- parameters (mostly vars now)
- dependents or impulses (vars_impulse or vars)
- independents / lagged dependents / responses (vars_response)
Hyperparameters only appear on their own, but for irfs and parameters we want to make both available (e.g. plot all lags of variable 1).
Also we should really make the names in plots nicer (than dep1-ind-1). Done in 2ea173e.
from bvar.
- Note that for parameters the variables seem to be switched (
vars_response = 1
plots all explanatories for dependent 1). Fixed in 2ea173e. - Also note that we might want to implement something like this for the summary method.
from bvar.
This can probably be tackled by just adjusting prep_data()
and changing / deprecating arguments.
from bvar.
Okay, so I am still not completely sure about the naming - atm I'm thinking of keeping the structure with response / impulse for IRFs and explicit subsetting of betas. But considering the vars argument in forecasts I want to allow users to pass both betas (dependents) and hyperparameters to the argument. As mentioned, this can probably be done in prep_data()
exclusively.
This then means that concurrent plot()
/ as.mcmc()
of hyperparameters and betas is possible.
Implementation-wise it's then probably easiest to switch from adding requested vectors (as in the beta-case atm) to prepping everything (as in the hyperpar-case) and then subset.
from bvar.
I think this was wrong:
Note that for parameters the variables seem to be switched (vars_response = 1 plots all explanatories for dependent 1). Fixed in 2ea173e.
from bvar.
So f3e639b fixes up a lot and includes some testing.
- The order of the data and names returned by
prep_data
definitely seems to be correct now. vars
now works via grep -- default is all hyperparameters. Hyperpars and dependent variables are checked directly with grep, dependents may also be accessed via an integer (i.e. position), independents also work via grep but need to specify a certain lag (e.g. ".*lag2" or "UNR.*lag4"), in order to be able to distinguish them from the independents.vars_impulse
andvars_response
are currently not functional and will need to be reimplemented and tested for compatibility (in density, plot (parameters, irfs and fcasts) and as.mcmc).
from bvar.
Looking good at the moment. I re-added vars_impulse
and vars_response
- they now overwrite whatever vars
does. I tested the examples and a bit more and it seems to work. Only thing that has to change is the example and function using independent_index()
.
See 2bb8010.
from bvar.
So by 4a48780 this seems to be working nicely. Nothing about independent_index()
needs to be changed, since e.g. vars_impulse
now feeds through get_var_set()
, meaning that one can either use grepping or positions (it's sufficiently smart about that).
The one update missing is the documentation.
from bvar.
So apart from the documentation this is done and look to be working nicely. (Note that I removed fixed = TRUE
to allow full grep in most cases.)
from bvar.
Related Issues (20)
- How to pass exogenous variables in the function bvar HOT 8
- NA resulting from log? HOT 6
- summary bvar HOT 2
- Support for mixed frequencies HOT 1
- FEVD plot HOT 10
- Questions regarding parallel implementation of BVAR HOT 4
- IRF colour HOT 2
- Overshrinkage with automatic psi HOT 1
- How to select the number of draws in fevd HOT 3
- Set prior mean automatically HOT 1
- Historical Variance Decomposition HOT 1
- narrative sign restrictions and anticipated vs unanticipated shock
- Diagnostics HOT 1
- Sign-restrictions HOT 12
- Default behavior for posterior summaries (mean over quantiles) HOT 2
- Noise added to unconditional forecast HOT 2
- Hyperparameters for monthly data? HOT 1
- Analyzing FEVD data HOT 2
- HDecomp plot
- Bind multiple chains
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from bvar.