dynare issueshttps://git.dynare.org/Dynare/dynare/-/issues2022-03-30T09:54:26Zhttps://git.dynare.org/Dynare/dynare/-/issues/1849Make default value of mh_jscale dependent on number of estimated parameters2022-03-30T09:54:26ZJohannes PfeiferMake default value of mh_jscale dependent on number of estimated parametersWe currently employ the default `mh_jscale=0.2`. This hardly ever fits as the optimal value is typically decreasing in the number of estimated parameters. I would propose to use the optimal value of Gelman et al. (1995) for the normal s...We currently employ the default `mh_jscale=0.2`. This hardly ever fits as the optimal value is typically decreasing in the number of estimated parameters. I would propose to use the optimal value of Gelman et al. (1995) for the normal symmetric random walk Metropolis-Hastings of `2.38^2/npar` as the default. That would break backward compatibility on files where the default was used, but would employ a more sensible starting value depending on the number of parameters estimated.6.xJohannes PfeiferJohannes Pfeiferhttps://git.dynare.org/Dynare/dynare/-/issues/1847Expand *_calibration to higher order2022-03-24T12:08:32ZJohannes PfeiferExpand *_calibration to higher orderCurrently, we only support first order. See also https://forum.dynare.org/t/gsa-of-irfs-with-a-higher-order-approximation/19758/2Currently, we only support first order. See also https://forum.dynare.org/t/gsa-of-irfs-with-a-higher-order-approximation/19758/26.xhttps://git.dynare.org/Dynare/dynare/-/issues/1801Allow estimation of parameters appearing in the discount factor2021-07-22T13:33:22ZJohannes PfeiferAllow estimation of parameters appearing in the discount factorRemaining task from https://git.dynare.org/Dynare/dynare/-/issues/1173Remaining task from https://git.dynare.org/Dynare/dynare/-/issues/11736.xhttps://git.dynare.org/Dynare/dynare/-/issues/1784Output updated_covariance from smoother2021-08-19T08:15:50ZMarco RattoOutput updated_covariance from smoothermore on 551917581fba7858f8f590c508b791716a1ace37, I have a question:
- we are issuing in `oo_.Smoother.Variance` the one step ahead variance of variables `V(t+1|t)`. this is in principle a duplicate of the `FilteredVariablesKStepAheadVar...more on 551917581fba7858f8f590c508b791716a1ace37, I have a question:
- we are issuing in `oo_.Smoother.Variance` the one step ahead variance of variables `V(t+1|t)`. this is in principle a duplicate of the `FilteredVariablesKStepAheadVariances`, for k=1.
- we are issuing in `oo_.Smoother.State_uncertainty` the smoother covariances `V(t|T)`
- we are not issuing the updated covariances `V(t|t)`: should we also provide this?
`V(t|t)` would be available already from `matlab/missing_DiffuseKalmanSmootherH3_Z.m` (`P` variable), while for `matlab/missing_DiffuseKalmanSmootherH1_Z.m` it is not (should be computed on purpose).6.xhttps://git.dynare.org/Dynare/dynare/-/issues/1766Allow for Herbst-Schorfheide and DS-MH sampler2021-07-21T21:12:52ZJohannes PfeiferAllow for Herbst-Schorfheide and DS-MH samplerSee https://forum.dynare.org/t/new-samplers-and-mode-compute-methods/17267/7
First part is done in https://git.dynare.org/Dynare/dynare/-/merge_requests/1884
- [ ] The default options for the samplers need to be added to `default_optio...See https://forum.dynare.org/t/new-samplers-and-mode-compute-methods/17267/7
First part is done in https://git.dynare.org/Dynare/dynare/-/merge_requests/1884
- [ ] The default options for the samplers need to be added to `default_option_values`
- [ ] The two routines currently do not provide any return arguments/output6.xhttps://git.dynare.org/Dynare/dynare/-/issues/1665Implement bridge sampler for computing marginal data density2021-07-20T10:57:50ZJohannes PfeiferImplement bridge sampler for computing marginal data density6.xhttps://git.dynare.org/Dynare/dynare/-/issues/1593Add truncated priors2021-09-01T14:33:18ZJohannes PfeiferAdd truncated priorsFollowing the discussions in #1591 and #750, we currently do not have a way to introduce a proper truncated prior. But this is an often requested feature that e.g. @rattoma would like to have. Waiting for the new estimation interface (e....Following the discussions in #1591 and #750, we currently do not have a way to introduce a proper truncated prior. But this is an often requested feature that e.g. @rattoma would like to have. Waiting for the new estimation interface (e.g. #824 and #846) seems to long to wait.
My specific proposal would be to add a new token `truncated_norm_pdf` that uses the third and fourth hyperparameter (which is usually reserved for generalized (beta) distributions) to specify the truncation. I consider this a natural interpretation of these hyperparameters for the normal distribution.
@stepan-a What do you think?
@rattoma Woud that satisfy your immediate needs? Or do you need a truncated distribution other than the normal?6.xStéphane Adjemianstepan@adjemian.euStéphane Adjemianstepan@adjemian.euhttps://git.dynare.org/Dynare/dynare/-/issues/1573Allow mode_compute to be a vector2021-08-15T19:07:59ZJohannes PfeiferAllow mode_compute to be a vectorAllow `mode_compute` to be a vector and the sequentially execute the mode-finders specified. See e.g. https://forum.dynare.org/t/simulated-annealing-block/11052 for why this may be useful.Allow `mode_compute` to be a vector and the sequentially execute the mode-finders specified. See e.g. https://forum.dynare.org/t/simulated-annealing-block/11052 for why this may be useful.6.xhttps://git.dynare.org/Dynare/dynare/-/issues/1556Allow running mode-finding on random draws from prior distribution to check f...2018-11-08T10:13:54ZJohannes PfeiferAllow running mode-finding on random draws from prior distribution to check for local modeshttps://git.dynare.org/Dynare/dynare/-/issues/1521create class for storing/writing metropolis draws2021-07-20T10:58:09ZHoutan Bastanicreate class for storing/writing metropolis drawsIn the `while` loop of `matlab/posterior_sampler_core.m` we have an example of how draws are stored and written when a certain number of draws have been stored in memory.
Need to create a Matlab class that stores:
- a vector
- a matrix
...In the `while` loop of `matlab/posterior_sampler_core.m` we have an example of how draws are stored and written when a certain number of draws have been stored in memory.
Need to create a Matlab class that stores:
- a vector
- a matrix
- a structure (`dr`)
And that writes itself to disk when a certain number of vector/matrix/structures have been written and clears itself so that more draws can be stored.
This class can then be used to standardize the various ways we do this throughout the Matlab codebase.6.xhttps://git.dynare.org/Dynare/dynare/-/issues/1514Add Importance Ratio as diagnostic for checking accuracy of normal approximat...2021-07-20T10:58:34ZJohannes PfeiferAdd Importance Ratio as diagnostic for checking accuracy of normal approximation to posteriorSee e.g. Slide 32 of http://apps.eui.eu/Personal/Canova/Teachingmaterial/bayes_dsge_eui2012.pdfSee e.g. Slide 32 of http://apps.eui.eu/Personal/Canova/Teachingmaterial/bayes_dsge_eui2012.pdf6.xhttps://git.dynare.org/Dynare/dynare/-/issues/1513Allow selecting proper training sample for endogenous_prior2021-07-20T11:00:59ZJohannes PfeiferAllow selecting proper training sample for endogenous_priorCurrently, we simply use `Y=data';`, but it is straightforward to include different dataCurrently, we simply use `Y=data';`, but it is straightforward to include different data6.xhttps://git.dynare.org/Dynare/dynare/-/issues/1170Non-bayesian estimation should use quasi-Maximum likelihood standard errors2021-07-27T11:37:26ZTom HoldenNon-bayesian estimation should use quasi-Maximum likelihood standard errorsAt present, with non-Bayesian estimation, Dynare computes standard errors using the Hessian of the likelihood. This is only valid if it is assumed that the shocks in the "true" model are normally distributed. And, in that case, it is an ...At present, with non-Bayesian estimation, Dynare computes standard errors using the Hessian of the likelihood. This is only valid if it is assumed that the shocks in the "true" model are normally distributed. And, in that case, it is an inefficient way of computing the standard errors, as it will be equal to the Fisher information matrix, which only requires the calculation of the derivative of the score vector.
It would make more sense to default to computing quasi-Maximum likelihood "sandwich" covariances, with the option to use the Fisher information matrix if the user wants quicker results.6.xMarco RattoMarco Rattohttps://git.dynare.org/Dynare/dynare/-/issues/1162Handling of trends2021-08-19T08:37:14ZMichelJuillardHandling of trendsCurrently trends are taken into account only if users indicate them for observed variables. However, in random walk with drift, a deterministic trend appears endogenously for variables that are not necessarily observed.
Let consider the...Currently trends are taken into account only if users indicate them for observed variables. However, in random walk with drift, a deterministic trend appears endogenously for variables that are not necessarily observed.
Let consider the following model where only Y is observed:
```
A_t = A_{t-1} + g + e_t
Y_t = A_t + eta_t
```
Both A_t and Y_t contain a deterministic linear trend with a slope of g. The current practice of only specifying the trend of Y is not satisfactory anymore.
When using the smoother, we need to recognize these trends and include them in SmoothedVariables and friends
https://git.dynare.org/Dynare/dynare/-/issues/1060Add capabilities for pseudo out of sample forecasting2019-11-21T08:36:44ZJohannes PfeiferAdd capabilities for pseudo out of sample forecastingAllow estimating model parameters on one part of the data and then, holding the estimated parameters fixed, forecast observations after the end of the sample used for estimation. See discussion on mailing list from September 14, 2015
Allow estimating model parameters on one part of the data and then, holding the estimated parameters fixed, forecast observations after the end of the sample used for estimation. See discussion on mailing list from September 14, 2015
https://git.dynare.org/Dynare/dynare/-/issues/933Add load_mh_file-like option for loading posterior subdraws2018-09-11T15:00:44ZJohannes PfeiferAdd load_mh_file-like option for loading posterior subdrawsSee #566
See #566
https://git.dynare.org/Dynare/dynare/-/issues/877document set_time, estimation_data, subsamples, joint_prior, prior, std_prior...2019-11-21T08:36:42ZHoutan Bastanidocument set_time, estimation_data, subsamples, joint_prior, prior, std_prior, corr_prior, prior equal, options, std_options, corr_options, options equalStéphane Adjemianstepan@adjemian.euStéphane Adjemianstepan@adjemian.euhttps://git.dynare.org/Dynare/dynare/-/issues/846Provide inverse gamma prior with indeterminate moments2021-09-01T14:31:35ZJohannes PfeiferProvide inverse gamma prior with indeterminate momentsWe currently only allow specifying inverse gamma priors with finite/unique mean and variance. But Leeper/Walker/Yang (2010) in an influential paper use an inverse gamma prior with s=1 and nu=4 (parametrization as at http://en.wikipedia.o...We currently only allow specifying inverse gamma priors with finite/unique mean and variance. But Leeper/Walker/Yang (2010) in an influential paper use an inverse gamma prior with s=1 and nu=4 (parametrization as at http://en.wikipedia.org/wiki/Inverse-gamma_distribution), which implies non-existing first and second moments. I would suggest to provide a new prior `inv_gamma_parametrized` that takes the values provided in the mean and standard deviation fields of the `estimated_params`-block directly as the parameters of the distribution, thereby avoiding the impossible transformation from mean and variances.
I would implement this when doing #5206.xhttps://git.dynare.org/Dynare/dynare/-/issues/824Add an interface for joint priors.2021-09-01T14:06:52ZStéphane Adjemianstepan@adjemian.euAdd an interface for joint priors.Only available for the new estimation syntax. Something like:
``` example
[alpha, beta].prior(shape=gaussian, mean=Vector, variance=Matrix, ...)
```
This interface is needed for Dirichlet priors over probabilities.
Only available for the new estimation syntax. Something like:
``` example
[alpha, beta].prior(shape=gaussian, mean=Vector, variance=Matrix, ...)
```
This interface is needed for Dirichlet priors over probabilities.
https://git.dynare.org/Dynare/dynare/-/issues/642Support Dirichlet prior distribution2021-09-03T12:01:06ZHoutan BastaniSupport Dirichlet prior distributionAdding dirichlet to the prior shape option for #568. Needs to be supported in estimation as well.
Adding dirichlet to the prior shape option for #568. Needs to be supported in estimation as well.