Thanks George. I guess that we can make a decision when we get a complete description of your algorithm. I guess examples that would cover typical case/use of partial information would be also helpful.
Best
Michel
G. Perendia wrote:
Hi
I believe (but may be wrong) that the "short-cut" does not mean we need to do it "properly" later (unless e.g. we want to introduce explicitly the expectation operator into Dynare model syntax)
I think I was setting the auxiliary (which I refer to as proxy) variable and its equation for the contemporary expectations on similar lines to those for E_{t-1}y_t, except that I used also the contemporary expectation error e_t which I thought was necessary.
However, I have to say that the "short" solution so far still does not yet answer how to deal with what in partial information we tend to call "auxiliary" variables (but may be called proxy instead if you like) - i.e. any of the noisy, optional, additional observations in the observation vector which (so far) are not deemed to be part of the core model's either endogenous or its exogenous sets, such as consumer's satisfaction index (obsCSX) or RPI based inflation measure. They are related to one or more of the core model's endogenous variables by auxiliary linear equation(s) which are not part of the main, core model and not directly part of the core model transition state matrix, but part of the part-info extended state space observation equation set W, e.g.:
obsY_t = y_t - the usual observation of y_t, and, .... obsRPI_t = PI_t [+ ep_t] and obsCSX_t = a*C_t +b*W_t + c*y_t [+ ec_t ]
where a, b and c can be either estimated or precalibrated parameters and e*_t are [optional] observation errors.
Those "auxiliary" observations and their relations will also need to be specified somewhere in the model but that can be in the 2nd stage of implementation.
Best regards
George
----- Original Message ----- From: "Michel Juillard" Michel.Juillard@ens.fr To: dev@dynare.org Cc: "joe pearlman" j.pearlman@londonmet.ac.uk Sent: Monday, November 24, 2008 12:10 PM Subject: Re: [DynareDev] integratio of Partial information and currentexpectations into Dynare
Hi there,
the guiding principle should be to do the integration right from the start
and
commit the necessary resources rather than having to do it over later.
- to you have a write-up with the algebra that needs to be performed for
partial information inside Dynare? 2) we have first to decide whether it is best to treat expectations as
different
(auxiliary) endogenous variables or to define a 4th type.
- Currently, in full information, if we want E_{t-1} y_t, one needs to
create an
auxiliary variable and an auxiliary equation ey = y(+1) and use ey(-1) instead of E_{t-1} y_t. If we are to provide an explicit operator E{i}{y}, the preprocessor would
need
to create the auxiliary variable, add the equation to the model, and
perform
the substitution as needed. We would need to keep a table of these auxiliary variables, but it doesn't
seem
necessary or desirable to create a 4th type of variable because the rules
for
output would be the same as for regular endogenous variables [Currently, I have also doubt about the distinction endogenous/exogenous,
but
this is another story...] Are there reasons to believe this would be different with partial
information?
Best
Michel
Quoting "G. Perendia" george@perendia.orangehome.co.uk:
Dear Joe
I have been thinking about the integration of the Partial information
and
current expectations into Dynare as you asked me to and I have some
ideas but
no easy solutions.
As I mentioned, a shortcut solution may require additional proxy
variables
for contemporary expectations such as ETYT_t = Et[y_t].
I expect an additional expectation error equations will be required that should avoid determinacy issues with the additional variable and the
possible
problems when solved, e,g, for steady state, poss on the lines of:
y_t = ETYT_t + e_t
Using then an additional structure (e.g. set inside the M_ or options_ structures) that is mapping expectations to their bases on the lines of:
['YTT' 'Y_t'; ......]
the Part info solver will be able to identify which of the relations are "dummy" and then transform and solve the model for decision rule using
its
own solver.
This solution however may become rather tedious to model from the user perspective, that is, if at all plausible - so, please let me know what
you
think. It will also require some parsing of the model within PI module
to
map-in expectations.
Alternatively, the proper integration would require changing
start-to-end
Dynare pipeline, from the parser/prerpocessor to KF estimation. It may
then
require adding 4th type of variable: expectation in addition to the
existing:
predetermined, forward and static variables, and, thus, changes to the numerous internal structures and functions.
This will, however, probably introduce an additional burden on the
design and
development resources, rather more than initially envisaged.
Best regards
George
--
Dev mailing list Dev@dynare.org http://www.dynare.org/cgi-bin/mailman/listinfo/dev
Dev mailing list Dev@dynare.org http://www.dynare.org/cgi-bin/mailman/listinfo/dev