# Stochastic Optimization

I am working on a Data Panel project that is about macro volatility, taking into consideration some quality indicators. According to my model, indicators definitely determine macro variables. Data are from 1944 up to 2020 for 50 countries with solid macro variables, quality indicators and lots of dummies.

What I would need is a code/program that solves the stochastic optimization problem(Newton Ramsay or any other), creating a new variable, perhaps discrete and continuous for each time period and for each one of the indicators included, such that it would by optimization find the optimum value of each the indicator that brings the actual GDP to its potential GDP value estimated with the HP filter. In other worlds, to find the optimal value for each of the indicators separately, creating a new variable for each of them and in each period that minimizes the cyclical component.

In simple math setting, say for indicator1:

Find value x of indicator1 such that minimizes (GDP potential -GDP actual) and is equal(=) to zero. I hope this is clear to you. I am fine with any software, thought prefer Matlab Stata or R

Any help if appreciated

Thank you.

• Hi: This may not be true for everyone but I can't understand what you're looking for. it's best to write down a sample equation for one of the periods because this might add clarity. Jul 30, 2021 at 12:14
• I am having a macro panel from 1944 to 2020 for 50 countries with solid macro variables, dummies and some custom quality indicators I create as part of the project. According to my model, indicators definitely determine macro variables. I am looking for a code/program to create a new variable continue and discrete for the indicator values for each period that brings the GDP to steady state. Find the value in every period i.e indicator1 s.t that value brings actual GDP to steady state(potential output),estimated with hp filter.The model is a state space with time variant coefficients Jul 30, 2021 at 22:12
• Your question has improved but maybe if you include the state space code, that might add more clarity ? I also am not clear on how you know when a macro variable has reached its steady state ? or why an indicator will bring it to a steady state ? Maybe others who are more economically inclined ( my background is not economics ) can help because, my guess is that, even if the clarity continues to improve, I still won't be able to help. Jul 31, 2021 at 23:35
• My model is a BVAR panel. I use BEAR toolbox,.Not relevant here. The steady state of macro variable is estimated with the hodrick Prescott filter. My theory suggests that the indicator determine the Macro variable in some way and there is a value of the indicator in each period that brings the Macro variable to the steady state. I am looking for a code or help on how to find those values of the indicator. Aug 1, 2021 at 20:35
• Hi: That sounds interesting but complex. Given that you're using a toolbox, I wouldn't have the slightest idea how to achieve your goal. Actually, even if you weren't using a toolobox, I wouldn't know how to achieve your goal either. Hopefully someone who is more familiar with BVAR panel models can chime in and add something. I'm sorry that I can't help. Aug 2, 2021 at 23:10