I am working on the construction of business cycle turning point indicator extrapolating the hidden signals from survey data.
I would like just some help in understanding if I am proceeding correctly
Basically, I want to use the realization of the observed survey data to get information on the hidden state of the economy (2 states “bad” or “positive”)
I assume that survey data follow a kind of autoregressive structure plus an innovation factor. That is
Panelists when participate to the survey provide their answer both referring to what has happened in the past (the AR component) plus some new elements.
These news are the factors that, in my view, can affect the regime transition probability
My Idea therefore is to use as the dependent variable the survey data and as switching regressor, the constant. The non-switching components are the AR part of equation 1
As the switching probabilities are affected by the information set, these news elements ( Et or better their estimates) are elements able to affect the probability regressors. Therefore I used them in the corresponding probability regressors box of the Equation estimate window (lagged by one observation)
Am I proceeding correctly in your view?
Thanks in advance for your help
Clemente De Lucia
For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.
1 post • Page 1 of 1
Who is online
Users browsing this forum: No registered users and 26 guests