A nonstationary model for output gap with a const in drift

General DSGE Modeling
1

Dear all,

I want to estimate a gap-model in levels with a stochastic trend that includes a constant in the drift, particularly:

% trend
mmu	= mmu(-1)  +  bbeta(-1)   + e_lev;
% drift
bbeta		= 	 rrho*bbeta(-1) + (1-rrho)*bbeta_const 	+ e_slp;
% cycle
 car		=   pphi1*car(-1) + pphi2*car(-2)	+ e_cyc;
% measurement
ym1		=    mmu(-1)  +	car(-1)	+ e_irr;
trend =  mmu(-1);
cycle = car(-1);

Note that I have no problem with the version where the drift is a random walk without a drift, i.e., rrho=1. However, I want to have more control over the trend growth rate and thus work with the version where 0<=rrho<1. I have this version in Eviews up and running. But I could not get Dynare run it. I use diffuse_filter option and tried steady(nocheck) with various initial values but Dynare spits errors about computing the initial likelihood and/or steady states.

The code attached.
gap_model.zip (88.4 KB)

2

The problem is that Dynare tries to approximate around the steady state, which does not exist if there is a drift. But as the model is already linear, the approximation point does not really matter. For that reason, you cannot use an

initval

block, which gives just the starting values for the steady state finding. Rather, you need to provide a “steady state” via the

steady_state_model

block and make sure that the correctness of the “steady state” is not checked via the nocheck option. The diffuse filter is needed because of the initial conditions when you use a unit root process. I tried

%  gin2017 

%----------------------------------------------------------------
% 1. Model variables and parameters
%----------------------------------------------------------------

var		bbeta		
		mmu		
		car
		ym1
        trend
        cycle;

varexo		e_irr		
		e_lev		
		e_slp		
		e_cyc		
        ;

parameters	
        rrho	
        bbeta_const
		pphi1
        pphi2	
        sig_e_irr sig_e_lev sig_e_slp sig_e_cyc 
        ;


%----------------------------------------------------------------
% 2. Calibration
%----------------------------------------------------------------

rrho		= 0;
bbeta_const= 0.0075;
pphi1		= 1.100;
pphi2		= -0.300;

% std of shocks
sig_e_irr=  0.003;
sig_e_lev=  .23; %.5;
sig_e_slp=  0; % 5
sig_e_cyc=   1; % normalize

%----------------------------------------------------------------
% 3. Model
%----------------------------------------------------------------

model(linear);


%%%%% STOCHASTIC TREND AND SLOPE %%%%%

mmu	= mmu(-1)  +  bbeta(-1)   + e_lev/1000 ;
bbeta		= 	 rrho*bbeta(-1) + (1-rrho)*bbeta_const 	+ e_slp/100;
car		=   pphi1*car(-1) + pphi2*car(-2)	+ e_cyc;

%%%%% MEASUREMENT EQUATION %%%%%

ym1		=    mmu(-1)  +	car(-1)	+ e_irr;
trend =  mmu(-1);
cycle = car(-1);
end;


steady_state_model;
mmu = 7.8;
ym1=7.8;
bbeta = bbeta_const;
car = 0;
trend = 7.8;
cycle=0;
end;

steady(nocheck);
 

%  initval;
%  mmu = 7.75;
% // bbeta = 0; 
% car = 0.03;
%  end;



%check; 
 
shocks;

var e_irr = sig_e_irr^2;
var e_lev = sig_e_lev^2;
var e_slp=sig_e_slp^2;
var e_cyc=sig_e_cyc^2;

end;


%----------------------------------------------------------------
% 4. Estimated parameters and data
%----------------------------------------------------------------
%%%%% PARAMETERS %%%%%
estimated_params;
 	rrho, .9359,   beta_pdf,       0.85, 0.075;
  	pphi1,	.2,	beta_pdf,       0.2, 0.075;
  	pphi2,	.2,	beta_pdf,       0.2, 0.075;

%%%%% INNOVATIONS %%%%%

 stderr	e_irr, .38,  	inv_gamma_pdf,  0.5, inf;
 stderr	e_lev,	.5,	inv_gamma_pdf,  0.5, inf;
 stderr	e_slp,	.5,	inv_gamma_pdf,  0.5, inf;
 stderr	e_cyc,	.5,	inv_gamma_pdf,  .5, inf;
end;

%
% Observed variables
%

varobs ym1;		

%----------------------------------------------------------------
% 5. Bayesian estimation and forecasting
%----------------------------------------------------------------
options_.console_mode=1; %(default: 0)

estimation	(datafile = lv_gdp_data, graph_format = pdf,
		nodiagnostic, diffuse_filter, 
        mh_drop=.5,  mh_jscale=0.2, 
        mh_replic = 0, 
        mh_nblocks = 2, filtered_vars, smoother,
		mode_compute = 4, plot_priors = 1,
		forecast = 0
		) trend cycle;

%----------------------------------------------------------------
% 6. Reporting
%----------------------------------------------------------------

verbatim;

lv_gdp_data;
% if laplace only
% smoothed
eval( 'trend_ 	= oo_.SmoothedVariables.trend'   ';' ]);
eval( 'cycle_ 	= oo_.SmoothedVariables.cycle'   ';' ]);
eval( 'irreg_ 	= oo_.SmoothedShocks.e_irr' ';' ]);

% %filtered
% eval( 'trend_f 	= oo_.FilteredVariables.trend'   ';' ]);
% eval( 'cycle_f 	= oo_.FilteredVariables.cycle'   ';' ]);
% trend_f(trend_f==0)=nan; 
% cycle_f(cycle_f==0)=nan;

figure(1)
plot(ym1,'black')
hold on;
plot(trend_,'red')
hold off;
figure(2)
plot(cycle_)

end;

and it seems to work OK

3

Danke, again.

4

In this case (a linear, nonstationary gap-model), what would be the easiest way to initialize the Kalman filter at specific starting values of endogenous variables different from their steady states? I think of a gap-process whose ss is zero but it can be far from the ss at the first observation. It seems adding the initval; block after the steady_state_model; block has no effect. The code attached.

[code]var
ym1
trend
cycle;

varexo
e_tr
e_cyc
;

parameters
drift
phi
sig_e_tr
sig_e_cyc
;

drift = 3/4; % 3 percent per year
phi = 0.7;
% std of shocks
sig_e_tr= .5;
sig_e_cyc= 1;

model(linear);
% trend
trend = trend(-1) + drift + e_tr;
% cycle
cycle = phicycle(-1) + e_cyc;
% measurement eq-n
100ym1 = trend + cycle;
end;

steady_state_model;
trend = 778.9; %775.9; %778.9;
cycle = 0; %3; % 0;
ym1 = 7.7890;
end;

steady(nocheck);

initval;
trend = 775.9;
cycle = 3;
ym1 = 7.7890;
end;

shocks;
var e_tr = sig_e_tr^2;
var e_cyc=sig_e_cyc^2;
end;

% estimated_params;

% phi, .7, beta_pdf, 0.7, 0.075;

% stderr e_tr, .5, inv_gamma_pdf, 0.5, inf;
% stderr e_cyc, 1, inv_gamma_pdf, 1, inf;

% end;

varobs ym1;

options_.console_mode=1; %(default: 0)

estimation (datafile = lv_gdp_data, nodisplay, graph_format = pdf,
nodiagnostic, diffuse_filter,
% kalman_algo = 1,
% lik_init=2,
mh_drop=.5, mh_jscale=0.2,
mh_replic = 0,
mh_nblocks = 2, filtered_vars, smoother,
mode_compute = 0, plot_priors = 1,
forecast = 0
) trend cycle;[/code]
gap_model.zip (88.1 KB)

5

With the diffuse_filter the diffuse initialization should take care of the fact that you can be far away from the steady state as the forecast error in the first period does not enter the likelihood. You can see this when you change the steady state value for ym1. This will change the deviation of the data from the steady state in the first period, but will not affect the likelihood.

6

The one-side filtered gap (oo_.FilteredVariables.cycle) indicates that Dynare initializes the gap at zero, whereas I want another specific initial point for the gap (e.g. a 3%) while having its steady state at zero.

Changing steady states for the trending variables does not help. E.g. if

steady_state_model; trend = 775.9; cycle = 0; ym1 = 7.7890; end;
then the initial point of the trend is 3% lower than the observable, but it also stays about 3% lower over the whole sample, meaning that the implied filtered gap (the observed variable minus the filtered trend) is shifted upwards over the whole sample.

If the steady states for both the trend and ym1 are lowered by 3%:

steady_state_model; trend = 775.9; cycle = 0; ym1 = 7.7590; end;
then the first observation of the implied filtered gap (the observed variable minus the filtered trend) is 3 but it vanishes to zero already in the second observation.

Also, in both cases the initial observation of the directly filtered gap (oo_.FilteredVariables.cycle) is zero and its filtered values differ by a lot from the implied filtered gap (the observed variable minus the filtered trend) over the whole sample, which to me is weird, as I expect both the directly and indirectly filtered gaps to coincide.

7

Are you sure you want to look at filtered variables, i.e. E_t(y_t+1)? The decomposition you have in mind seems to rely on updated variables E_t(y_t). **

8

Pardon, updated variables. So, please ignore the last point - the two are the same.

But otherwise the issue remains - the updated gap starts at zero unless its steady state is nonzero. However, a practitioner often knows where the gap should be at the start of the sample and thus wants to direct the Kalman filter by setting initial values. So, is there a way to initialize the gap at a specific value other than its zero steady state? I wouldn’t bother if it wasn’t so easy to achieve in Eviews.

9

Dear Ginters,
I am working on it and will keep you updated.

10

Hi,
This topic is of interest to me. Is there any update on the issue of starting values as raised by Ginters?
Thanks.

11

In the unstable version (soon Dynare 4.7), there is the filter_initial_state-block.