Parameter representation in endogenous MS models in RISE

[pengsimin77]

Dear Junior Maih, I hope you are well.

I am working with an endogenous Markov-switching DSGE model in RISE and had a question regarding parameter representation.

In my .rs file, I specify regime-specific monetary policy coefficients
phi_pi(coef,1), phi_y(coef,1), phi_pi(coef,2), and phi_y(coef,2).

However, after running
m = rise('EG_MS_DSGE_est','rise_flags',struct('exogenous_switching',false));

these coefficients do not appear in m.parameters. Instead, I obtain parameters such as
iota_1___, iota_2___, part_phi_pi, part_phi_y, phi_pi, phi_y, sig___, ss_phi_pi, ss_phi_y.

As a result, I can only estimate these parameters, rather than the regime-specific policy coefficients directly. I was wondering:

• What is the interpretation of these parameters in the endogenous switching setup?
• How are the regime-specific monetary policy coefficients constructed from them?
• More generally, is there a recommended way to extract the numerical values of model parameters in RISE? I noticed that m.parameters does not seem to store the parameter values directly.

For your convenience, I have attached my .rs file; it was saved as a .txt file for easier upload.

EG_MS_DSGE_est.txt

Thank you very much for your time and for developing RISE.

Best regards,
Simin

[pengsimin77]

I believe I have solved part of the issues after reviewing your new repository. I can set priors for these coefficients by names phi_pi_coef_1, phi_pi_coef_2, ect. And I can retrieve the estimated values using get(m_est, 'mode'); . However, I am unsure how to extract the values of the other parameters that I set in the RS file.

[jmaih]

get(m,'parameters') returns the parameter values for all parameters in a structure.

…

On Fri, Jan 23, 2026 at 12:15 PM pengsimin77 ***@***.***> wrote: *pengsimin77* left a comment (jmaih/RISE_toolbox#229) <#229 (comment)> I believe I have solved part of the issues after reviewing your new repository. I can set priors for these coefficients by names phi_pi_coef_1, phi_pi_coef_2, ect. And I can retrieve the estimated values using get(m_est, 'mode'); . However, I am unsure how to extract the values of the other parameters that I set in the RS file. — Reply to this email directly, view it on GitHub <#229 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AATKBT2FHGCEMBFCP5OKA734IH7EHAVCNFSM6AAAAACSRCYIZ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTOOBZG43TANJUGU> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[pengsimin77]

Dear Professor Maih,

Thank you very much for your reply — it was very helpful.

It seems that in RISE, the get(m, '...') and set(m, '...', ...) syntax is quite powerful for accessing and modifying model features. I have learned that entries such as 'data', 'priors', 'mode', and 'parameters' can be handled in this way.

May I ask if there is a systematic way to inspect or list all available fields/keywords that can be accessed or modified via get and set? For example, is there a function, documentation page, or internal structure that enumerates all admissible entries?

Thank you again for your time and help.

Best regards,
Simin

[jmaih]

Dear Simin, For the version of RISE you are using, below is a complete list of the admissible options supported by the `get` and `set` commands, together with a brief description of what each option returns or modifies. ================================ GETTABLE OPTIONS ================================ *- 'structure' * Returns model derivatives, transition matrices, and auxiliary information required to solve the model. *- 'definitions'* Returns the evaluated model definitions (elements that enter the model block with a # sign). *- 'solution'* *- 'bgp' * Returns the balanced growth path (BGP). Variants: - 'bgp' or 'bgp(default)' - 'bgp(struct)' - 'bgp(cell)' *- 'sstate'* Returns the steady state. Variants: - 'sstate' or 'sstate(default)' - 'sstate(struct)' - 'sstate(cell)' *- 'parameters' OR 'par_vals'* Returns parameter values. Variants: - default - '(struct)' - '(cell)' *- 'par_list' * Returns the list of parameters. Sub-options: - '(switching)' - '(trans_prob)' - '(measurement_error)' - '(in_use)' *- 'par_tex'* Returns LaTeX/descriptive names of parameters. *- 'endo_list' * Returns the list of endogenous variables. Sub-options: - '(lagrange_multiplier)' - '(static)' - '(predetermined)' - '(pred_frwrd_looking)' - '(state)' - '(frwrd_looking)' - '(log_var)' - '(log_expanded)' - '(auxiliary)' - '(original)' - '(affect_trans_probs)' - '(hybrid_expect)' - '(stationary)' *- 'endo_tex' * Returns LaTeX/descriptive names of endogenous variables. *- 'exo_list' * Returns the list of exogenous variables. Sub-options: - '(observed)' - '(in_user)' *- 'exo_tex' * Returns LaTeX/descriptive names of exogenous variables. *- 'obs_list' * Returns the list of observable variables. Sub-options: - '(endogenous)' *- 'obs_tex' * Returns LaTeX/descriptive names of observable variables. *- 'def_list' * Returns the list of model definitions. *- 'chain_list'* Returns the list of Markov chains. *- 'chain_tex'* Returns descriptions of Markov chains. *- 'state_list' * Returns the list of states across all Markov chains. *- 'state_tex'* Returns descriptions of Markov-chain states. *- 'regime_list' * Returns the list of regimes (combinations of states). *- 'regime_tex'* Returns descriptions of regimes. *- 'tex' or 'description'* Returns descriptions of all model atoms (parameters, variables, chains, states). Sub-options: - '(math)' - '(long)' *- 'state_vars'* Returns state variables and their lag structure used for forecasting. *- 'mode'* Returns posterior mode of estimated parameters. *- 'prior_mean'* Returns prior means of estimated parameters. *- 'start'* Returns starting values used in estimation. *- '<parameter_name>' * Returns values of a specific parameter by name. ================================ SETTABLE OPTIONS ================================ GENERAL OPTIONS *- 'parameters' * Sets parameter values. *- 'tex_name'* Sets LaTeX/descriptive names for parameters, variables, chains, or states. *- 'legend' * Sets a legend name for the model. *- 'optimset'* Sets optimization options (passed to numerical optimizers). *- 'results_folder'* Sets the folder where results are stored. DATA AND ESTIMATION OPTIONS *- 'data'* Sets the dataset used for estimation. *- 'data_cond_ct' * Sets conditional central paths for data. *- 'data_cond_lb' * Sets lower bounds for conditional data paths. *- 'data_cond_ub'* Sets upper bounds for conditional data paths. *- 'estim_start_date' * Sets estimation start date. *- 'estim_end_date' * Sets estimation end date. *- 'data_demean' * Toggles demeaning of data. *- 'kf_presample'* Sets presample length for Kalman filtering. PRIORS AND RESTRICTIONS *- 'priors' / 'estim_priors'* Sets parameter priors. *- 'estim_endogenous_priors'* Sets priors on the behavior of the model. *- 'estim_linear_restrictions' * Sets linear parameter restrictions. *- 'estim_nonlinear_restrictions' * Sets nonlinear parameter restrictions. *- 'estim_general_restrictions' * Sets general estimation restrictions. SOLUTION AND ROUTINES *- 'solve_shock_horizon' * Sets agents’ anticipation horizon for shocks (possibly regime-dependent). *- 'solve_function_mode'* Selects function handling mode: 'explicit' / 'amateur', 'vectorized' / 'professional', or 'disc'. *- 'solve_perturbation_type' * Sets perturbation type and triggers auxiliary parameter updates. FORCED OPTIONS *- '+option_name' * Forces the insertion of a custom option into the model options structure. If you would like, I can also provide the same lists in table form or split by task (solution, estimation, simulation). Best regards, Junior

…

On Sat, Jan 24, 2026 at 2:26 PM pengsimin77 ***@***.***> wrote: *pengsimin77* left a comment (jmaih/RISE_toolbox#229) <#229 (comment)> Dear Professor Maih, Thank you very much for your reply — it was very helpful. It seems that in RISE, the get(m, '...') and set(m, '...', ...) syntax is quite powerful for accessing and modifying model features. I have learned that entries such as 'data', 'priors', 'mode', and 'parameters' can be handled in this way. May I ask if there is a systematic way to inspect or list all available fields/keywords that can be accessed or modified via get and set? For example, is there a function, documentation page, or internal structure that enumerates all admissible entries? Thank you again for your time and help. Best regards, Simin — Reply to this email directly, view it on GitHub <#229 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AATKBTZ6FROJG6Y3WHHTTFD4INXKDAVCNFSM6AAAAACSRCYIZ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTOOJUGYZDSNRVGY> . You are receiving this because you commented.Message ID: ***@***.***>

[pengsimin77]

Dear Professor Maih,

Thank you very much for your reply — I learned a great deal from it. If it is convenient for you, I would be grateful to have the lists in table form or organized by task (solution, estimation, simulation), as this would greatly help my understanding.

Best regards,
Simin