## placing restrictions on coefficients sum to one

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

Sanjana.R
Posts: 2
Joined: Thu Oct 24, 2013 12:01 pm

### placing restrictions on coefficients sum to one

Hi there,

I am quite new to eviews and I have been trying to estimate a model for my thesis. This model is a PHillips Curve model and I want to put restrictions on previous π_t sum to one.

For more clarification I am posting ym model here.#

π_t=1/4 ( π_(t-1)+π_(t-2)+π_(t-3)+π_(t-4) )+α(u-u*)t+ε_t

where π_t is annualized quarterly inflation rate
1/4 ( π_(t-1)+π_(t-2)+π_(t-3)+π_(t-4) is expected inflation ( expected inflation being an average og past four quarters )

I want my restriction to be that the coefficient of expected inflation to be sum to 1.

startz
Non-normality and collinearity are NOT problems!
Posts: 3507
Joined: Wed Sep 17, 2008 2:25 pm

### Re: placing restrictions on coefficients sum to one

Code: Select all

`ls pi=c(1) + c(2)*pi(-1) + c(3)*pi(-2) + c(4)*pi(-3) + (1-c(2)-c(3)-c(4))*pi(-4)`

Sanjana.R
Posts: 2
Joined: Thu Oct 24, 2013 12:01 pm

### Re: placing restrictions on coefficients sum to one

but how does that work?

Can I assumer the first set of pi-s to be a simple moving average model based on past four quarters? and then do the regression?

Giansonn
Posts: 1
Joined: Sat Mar 17, 2018 5:19 am

### Re: placing restrictions on coefficients sum to one

Yes and how can we obtain the coefficient and s.e. of the coefficient thats missing?

startz
Non-normality and collinearity are NOT problems!
Posts: 3507
Joined: Wed Sep 17, 2008 2:25 pm

### Re: placing restrictions on coefficients sum to one

Do a Wald test on 1-c(2)-c(3)-c(4)=0