Thank you very much for all the useful videos. I have a question regarding the extended path. Is it equivalent to use extended_path command to use perfect_foresight_solver sequentially? I mean could I simulate a shock using perfect_foresight_solver in period one, take the value of the endogenous variables in period two as starting values for the perfect_foresight_solver and simulate another shock. If I do so for lets say 100 periods with random shocks that would be equivalent to extended_path?
Thanks a lot for sharing! Big fan of yours! At 18:10, you say y is simulated, may I know how is it different from the y_eta_a, y_eta_b, What does it mean by y is drawn from 1000 shocks, may I have some examples of this? is it because period = 1000? Is period = 1000 mean there is a shock at each period for 1000 periods or 1000 shocks at each period for 1000 periods? What does period = 1000 mean? Is this value y shown in the irfs, where is it used for? Is y_eta_a the irf of y for eta_a, is irf only a one-time shock for 1 period? Then how is the stochastic situation different from the deterministic situation in irfs
y_eta_a and y_eta_b are the Impulse Response functions with respect to either eta_a or eta_b. That is, there is a single shock in period 1 and no shocks afterwards and we plot for 20 periods (option irf=20) the trajectory of y. Just y is the trajectory of y for 1000 periods (periods=1000), but here we not only looked at one individual shock in the impact period, but at ALL shocks at ALL periods. So you draw 1000 different values for ALL the shocks in the model from the normal distribution and plug those into the policy function and then plot the trajectory.
Thank you so much for your video， it really helps me a lot! I'm currently working on a paper about calculating the natural rate of interest by a DSGE model, and I really don't know how to work it out in Dynare. I was wondering if you could upload some guidance about the calculation of the natural rate in DSGE, thank you so much, Professor!
Thank you for this. Is it possible to include a max or min function in the stochastic simulation as well? Similar to how you handled the lower bound for the interest rate?
No, for this you would need to use the occBin Toolbox of Dynare at the caveat that it only works with first-order perturbation methods. Have a look at Day 1 materials here: git.dynare.org/Dynare/dynare/-/wikis/Dynare-Workshop-for-Advanced-Users-2022
What specifically are you interested in? There are good introductions to this concept on TH-cam already. I try to focus on the parts which are relevant for my courses and honestly for macroeconomists a deep mathematical understanding of dynamic programming is desirable but often it is sufficient to have a general idea and just know how to apply it.
Thanks a lot for the brief introduction!! Very helpful for somebody who just got thrown into the cold water to esse Dynare :D.
Don't forget to have fun with it!
Thank you very much for all the useful videos.
I have a question regarding the extended path.
Is it equivalent to use extended_path command to use perfect_foresight_solver sequentially?
I mean could I simulate a shock using perfect_foresight_solver in period one, take the value of the endogenous variables in period two as starting values for the perfect_foresight_solver and simulate another shock.
If I do so for lets say 100 periods with random shocks that would be equivalent to extended_path?
Thanks a lot for sharing! Big fan of yours!
At 18:10, you say y is simulated, may I know how is it different from the y_eta_a, y_eta_b,
What does it mean by y is drawn from 1000 shocks, may I have some examples of this? is it because period = 1000?
Is period = 1000 mean there is a shock at each period for 1000 periods or 1000 shocks at each period for 1000 periods? What does period = 1000 mean?
Is this value y shown in the irfs, where is it used for?
Is y_eta_a the irf of y for eta_a, is irf only a one-time shock for 1 period? Then how is the stochastic situation different from the deterministic situation in irfs
y_eta_a and y_eta_b are the Impulse Response functions with respect to either eta_a or eta_b. That is, there is a single shock in period 1 and no shocks afterwards and we plot for 20 periods (option irf=20) the trajectory of y. Just y is the trajectory of y for 1000 periods (periods=1000), but here we not only looked at one individual shock in the impact period, but at ALL shocks at ALL periods. So you draw 1000 different values for ALL the shocks in the model from the normal distribution and plug those into the policy function and then plot the trajectory.
Thank you so much for your video， it really helps me a lot! I'm currently working on a paper about calculating the natural rate of interest by a DSGE model, and I really don't know how to work it out in Dynare. I was wondering if you could upload some guidance about the calculation of the natural rate in DSGE, thank you so much, Professor!
Thank you very much
Where can I find the model you used, sir?
Sir, u have uploaded Advanced Macroeconomics Part 2 series. But, where is part 1. Can u please upload that too? Please Sir🙏🙏
That was a course at the University of Tübingen, part I is taught by another lecturer and has a different scope.
Thank you for this. Is it possible to include a max or min function in the stochastic simulation as well? Similar to how you handled the lower bound for the interest rate?
No, for this you would need to use the occBin Toolbox of Dynare at the caveat that it only works with first-order perturbation methods. Have a look at Day 1 materials here: git.dynare.org/Dynare/dynare/-/wikis/Dynare-Workshop-for-Advanced-Users-2022
@@wmutschl Thank you most kindly for the swift response. All the best!
Sir, can u please upload videos on dynamic programming and stochastic dynamic programming?
What specifically are you interested in? There are good introductions to this concept on TH-cam already. I try to focus on the parts which are relevant for my courses and honestly for macroeconomists a deep mathematical understanding of dynamic programming is desirable but often it is sufficient to have a general idea and just know how to apply it.