Estimation of Taylor rule with smoothing parameter

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.

Moderators: EViews Gareth, EViews Moderator

res
Posts: 5
Joined: Sun Apr 15, 2018 5:45 am

Estimation of Taylor rule with smoothing parameter

Postby res » Sun Apr 15, 2018 7:23 am

Hello,
I am interested in the estimation of Taylor rule with interest rate smoothing, like in the attached picture.
I have done only simple regressions, but I have never done regression with rho(smoothing parameter).
Can anyone suggest how it can be done in EViews?
Thank you in advance!
Image

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Estimation of Taylor rule with smoothing parameter

Postby startz » Sun Apr 15, 2018 7:47 am

You can specify a regression in EViews by writing out an equation using c(1), c(2), etc as names for the coefficients, as in

Code: Select all

ls y=c(1)*c(2) + c(2)*x

res
Posts: 5
Joined: Sun Apr 15, 2018 5:45 am

Re: Estimation of Taylor rule with smoothing parameter

Postby res » Sun Apr 15, 2018 9:43 am

Thank you for your reply.
I now understand that I porbably need to do this regression with GMM method, I specified everything, but something is wrong as I get message singular matrix. How can I fix this?
Attachments
Screenshot (2).png
Screenshot (2).png (10.63 KiB) Viewed 6246 times

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Estimation of Taylor rule with smoothing parameter

Postby startz » Sun Apr 15, 2018 9:53 am

I think that the problem is that you have both included c and clicked include a constant in the instrument list.

res
Posts: 5
Joined: Sun Apr 15, 2018 5:45 am

Re: Estimation of Taylor rule with smoothing parameter

Postby res » Sun Apr 15, 2018 10:00 am

I deleted c from the instrumental list, but it still does not work and I get message singular matrix. Maybe you have other ideas?
I am so desperate, don't know how to solve this problem.

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Estimation of Taylor rule with smoothing parameter

Postby startz » Sun Apr 15, 2018 10:55 am

What you've done looks right. Since your model is linear, you might try the TSLS command and see if that works. Also, before you run GMM set c=0.

res
Posts: 5
Joined: Sun Apr 15, 2018 5:45 am

Re: Estimation of Taylor rule with smoothing parameter

Postby res » Sun Apr 15, 2018 11:11 am

What do you mean by set c=0? How can I do that? Will it impact results?
Thank you in advance!

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Estimation of Taylor rule with smoothing parameter

Postby startz » Sun Apr 15, 2018 11:13 am

Type c=0 into the command line before estimating an equation. This resets the starting values of the coefficients. Not likely to matter, but there's a chance.

res
Posts: 5
Joined: Sun Apr 15, 2018 5:45 am

Re: Estimation of Taylor rule with smoothing parameter

Postby res » Sun Apr 15, 2018 11:30 am

It still does not work.
Tried to do the same way as in this pdf http://coin.wne.uw.edu.pl/~lgoczek/pdf/ ... trics5.pdf, but no results.
The problem only disappears when I remove a constant term c(2) from the regression, but I cannot use such result, I think.
Anyway, thank you for your help and time, I really appreciate that!


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 24 guests