Compatibility issue with dyntable

Stochastic simulations
1

Hi again. So originally, the error I was getting was:
There are 21 eigenvalue(s) larger than 1 in modulus
for 22 forward-looking variable(s)
The rank condition ISN’T verified!
Error using print_info (line 32)
Blanchard & Kahn conditions are not satisfied: indeterminacy.

The model seems to run (almost) without the BK errors, if I include either one of the lagrange multipliers (lambdah or lambdal), associated with either one of the two sectors’ borrowing constraints. If I include both I get the BK errors back as follows:
There are 21 eigenvalue(s) larger than 1 in modulus
for 20 forward-looking variable(s)
The rank condition ISN’T verified!
Error using print_info (line 32)
Blanchard & Kahn conditions are not satisfied: no stable equilibrium.

I have two questions

  1. is there any justification for the lagrange multiplier on the borrowing constraint of a sector to be predetermined? And why for only one sector when there is another symmetrically borrowing constrained sector?
  2. Even when I do not get the BK errors, once I allow one of the lagrange multipliers to be predetermined, the model runs all the way to variance decomposition and then gives the following errors:
    Error using cellfun
    Input #2 expected to be a cell array, was char instead.

Error in cellofchararraymaxlength (line 20)
if ~all(cellfun(@ischar, c))

Error in dyntable (line 57)
headers_length = cellofchararraymaxlength(headers(2:end));

Error in US_IC3_v2.driver (line 1176)
dyntable(options_,‘Relative standard deviations’,strvcat(‘VARIABLE’,‘REL. S.D.’),M_.endo_names(1:74,:),statistic1,10,8,4);

Error in dynare (line 281)
evalin(‘base’,[fname ‘.driver’]);

It seems to do with the dyntable command - any idea how to fix this one? I attach the modfile again. Thanks very much in advance. Any help will be greatly appreciated!
US_IC3_v2.mod (14.7 KB)

2
  1. I don’t know enough about your model to give a detailed answer. But generally, there is only one unique valid timing that you need to find. Sometimes it is possible to get the model to run by altering the timing of correctly time variable, but that would not result in correct results. So tread carefully. Regarding the two constraints: you should think carefully about needs to go into the model. If you drop one multiplier, you will also be leaving out one equation. Is the equation you leave out really redundant, e.g. due to market clearing?
  2. Since Dynare 4.6 the variables names need to be cell arrays:
dyntable(options_,'Relative standard deviations',{'VARIABLE','REL. S.D.'},M_.endo_names(1:74,:),statistic1,10,8,4);
3

Many thanks. It has finally worked. Ultimately if I allow household consumption to be determined using the household’s budget constraint (instead of from the aggregate resource constraint as I was doing) the mod file runs with simply the capital stocks as the predetermined variables. I don’t understand why that is yet since the steady state solutions for household consumption are identical both ways, but some inherent timing issue must be involved (need to think about it). And thanks for the updated dyntable command. That took care of the other error.