## Favar QUESTION

Moderators: EViews Gareth, EViews Moderator, EViews Esther

ouerk
Posts: 14
Joined: Tue May 23, 2017 11:29 am

### Re: Favar QUESTION

thank you , Dakila !!i am grateful ! we are lucky to be able to benefit from your help !

i followed your instructions and i obtain the impulse response matrix(for shock egal to -0.25) , i use the option graph and i obtain the graphs of impulse response without confidence intervals , the axis 0 and the name of the selected variables (see workfile). what should i do to get them (like figure 01 in your exple) ?

for the conditional forecast,you tell me i should transform for exple series2 to the non standardized on . i must do it :
-before i estimate eq01 ?
- before i forecast the factors ?
- before i forecast series2 ?

i transform only the selected variable (series2) , NOT the series contrained (ffr). true ?

THANK YOU very much !

Salima
Attachments
graph.wf1

dakila
Posts: 311
Joined: Tue Nov 24, 2015 4:57 pm

### Re: Favar QUESTION

what should i do to get them (like figure 01 in your exple) ?

You should do the same things for the confidence interval. For example:

Code: Select all

`matrix irf_ub=irfxmat_ub01/irfxmat_ub01(1,1)irf_ub = irf_ub*(-0.25/ffr_std) ' upper boundmatrix irf_lb=irfxmat_lb01/irfxmat_lb01(1,1)irf_lb = irf_lb*(-0.25/ffr_std) ' lower bound`

Easiest way to draw the figure is copy the matrix to excel then draw the graph.

for the conditional forecast,you tell me i should transform for exple series2 to the non standardized on . i must do it :
-before i estimate eq01 ?
- before i forecast the factors ?
- before i forecast series2 ?

After the conditional forecast you should transform for expel series2
i transform only the selected variable (series2) , NOT the series contrained (ffr). true ?

Yes that is true.

ouerk
Posts: 14
Joined: Tue May 23, 2017 11:29 am

### Re: Favar QUESTION

i obtain the ifr with confidence interval.

i transform some of my variables to induce stationarity .when i perform 'the conditional forecast ',i have for exple the conditional forecast of the
dlog pib ( i transform this variable back to the non-standardized one).

i would like to obtain the conditional forecast of the real value (without transformation ). in general ,i can applied the formula
Tx+1 = Tx *e(dlogpib) . how can i traduce this formula to employ it in eviews ?

SALIMA

dakila
Posts: 311
Joined: Tue Nov 24, 2015 4:57 pm

### Re: Favar QUESTION

what is Tx?

ouerk
Posts: 14
Joined: Tue May 23, 2017 11:29 am

### Re: Favar QUESTION

hi Dakila !

can you add the transformation' second difference of log ' and 'first difference' please ? i have no stationary variables with dlog so i must use ddlog.

Thank you very much !!

salima Ouerk
phd student
Last edited by ouerk on Wed Aug 23, 2017 4:41 am, edited 1 time in total.

dakila
Posts: 311
Joined: Tue Nov 24, 2015 4:57 pm

### Re: Favar QUESTION

I will try

ouerk
Posts: 14
Joined: Tue May 23, 2017 11:29 am

### Re: Favar QUESTION

thank you very much !!

ouerk
Posts: 14
Joined: Tue May 23, 2017 11:29 am

### Re: Favar QUESTION

Hi Dakila,

could you add the others possible transformations ? ( the fist difference ,the second difference and ddlog)

Salima Ouerk

ouerk
Posts: 14
Joined: Tue May 23, 2017 11:29 am

### Re: Favar QUESTION

Hi,

The impulses responses of the variables to an unexpected increase in ffr are to be interpreted in "pourcentage points " in the model ? for exple, changes in GDP are expressed in pourcentage point ? or in standard deviation ?

Thanks

dakila
Posts: 311
Joined: Tue Nov 24, 2015 4:57 pm

### Re: Favar QUESTION

Note that the figures report impulse responses in standard deviation units.