FAVAR add-in

For questions about EViews Add-ins available from the EViews Add-ins webpage. Note each add-in available on our webpage will have its own individual thread.

Moderators: EViews Gareth, EViews Moderator, EViews Esther

jjw_10
Posts: 4
Joined: Sat Dec 28, 2013 10:32 am

Re: FAVAR add-in

Postby jjw_10 » Wed Jun 07, 2017 8:24 am

Dakila - Thank you for the useful add-in. Is there a way to apply the scaled IRF add-in to the FAVAR add-in?

dakila
Posts: 479
Joined: Tue Nov 24, 2015 4:57 pm

Re: FAVAR add-in

Postby dakila » Wed Jun 07, 2017 4:35 pm

Good idea. I will try to code it.

jjw_10
Posts: 4
Joined: Sat Dec 28, 2013 10:32 am

Re: FAVAR add-in

Postby jjw_10 » Fri Jun 09, 2017 7:02 am

Thank you, Dakila. Is there a possibility of running cumulative IRFs in your add-in?

dakila
Posts: 479
Joined: Tue Nov 24, 2015 4:57 pm

Re: FAVAR add-in

Postby dakila » Fri Jun 09, 2017 5:54 pm

Yes, It actually do the cumulative IRFs when you transform the variable. For example, if you put the transformation code 4 or 5 it will accumulate the IRFs.

jjw_10
Posts: 4
Joined: Sat Dec 28, 2013 10:32 am

Re: FAVAR add-in

Postby jjw_10 » Fri Jun 09, 2017 8:08 pm

Ahh, ok. Thanks.

adriana
Posts: 2
Joined: Sat Jun 10, 2017 1:57 pm

Re: FAVAR add-in

Postby adriana » Thu Jun 22, 2017 6:09 pm

Hi,

I tried to get the factors by the usual way of principal components, but I didn´t have the same series. Also I can see that the final factors used in VAR have "rot" in its labels. Does it have any other treatment, or could you please tell me the way to get it? Thanks!

dakila
Posts: 479
Joined: Tue Nov 24, 2015 4:57 pm

Re: FAVAR add-in

Postby dakila » Fri Jun 23, 2017 2:29 am

"ROT" means the rotated factors. In order to estimate the factors do the following steps:
1) estimate factors (C) using all variables except FFR (you did this step)
2) estimate slow moving factors (Fs) from the slow moving variables.
3) estimate the regression by OLS : C=b1*Fs + b2*FFR + e
4) estimate the rotated factors : C-b2*FFR

adriana
Posts: 2
Joined: Sat Jun 10, 2017 1:57 pm

Re: FAVAR add-in

Postby adriana » Mon Jun 26, 2017 6:05 pm

Thank you! I got them but with a very small difference. Maybe that's because I didn't standardize my variables (since all of them were stationaries), do you think thats made the difference?

dakila
Posts: 479
Joined: Tue Nov 24, 2015 4:57 pm

Re: FAVAR add-in

Postby dakila » Mon Jun 26, 2017 7:19 pm

Yes, I think that made the difference.

ilef
Posts: 2
Joined: Fri Jun 30, 2017 6:23 am

Re: FAVAR add-in

Postby ilef » Fri Jun 30, 2017 7:29 am

Hello, I want to apply FAVAR on eviews but I found difficulties in entering variables on the dialog box, please can you help me?
Thanks a lot.

dakila
Posts: 479
Joined: Tue Nov 24, 2015 4:57 pm

Re: FAVAR add-in

Postby dakila » Fri Jun 30, 2017 4:44 pm

Did you read the instruction which is located in C:\Users\...\Documents\EViews Addins\FAVAR (Favar Package.pdf)?

ilef
Posts: 2
Joined: Fri Jun 30, 2017 6:23 am

Re: FAVAR add-in

Postby ilef » Sat Jul 01, 2017 2:53 pm

Yes but I still have problems!
Could you explain how to enter variables polease.
Thanks a lot.

dakila
Posts: 479
Joined: Tue Nov 24, 2015 4:57 pm

Re: FAVAR add-in

Postby dakila » Sat Jul 01, 2017 4:09 pm

there is an example program which replicates BBE (2005)

Code: Select all

' BBE (2005) replicaton
' Figure 2
' Table 1

%path = @runpath
cd %path

' open the BBE data that is transformed to induce stationarity
wfopen(type=txt) nsbalpanel.txt delim=space
pagestruct(freq=m,start=1959)

'rename some variables   

for !i=1 to 9
   rename series0{!i} series{!i}
next

'standardize all variables for principal component analsyis

for !i=1 to 120
   series{!i}=(series{!i} - @mean(series{!i}))/@stdevp(series{!i})
next

'Y variables 
genr ffr=series77

'All variables (X) except Y variable (FFR)

group xdata
for !i=1 to 120
   xdata.add series{!i}
next
xdata.drop series77

' Slow-moving variables
group xslow
for !i=1 to 53
   xslow.add series{!i}
next
for !i=103 to 119
   xslow.add series{!i}
next

' Selection of key variables for impulse response analysis
group xir
xir.add series16 series108 series78 series81 series96 series93 series74 series102 series17 series49 series50 series51 series26 series48 series118 series54 series62 series71 series120

'Vector of transformation code of Y (ffr) variable + selection of key variables (xir)
'1 for no transformation
'4 for logarithm
'5 for first difference of logarithm

vector tcode =@fill(1, 5, 5, 1, 1, 5, 5, 5, 1, 1, 5, 5, 5, 1, 1, 5, 4, 1, 1, 1)

'Name of Y variables Y(ffr) + selected key variables (xir)
' Please first put a name of Y variable then selected variables
text yx_name
yx_name.append(file) yxirlabel.txt

' or append manually
' yx_name.append FFR
' yx_name.append IP
' yx_name.append CPI
'......
' yx_name.append Consumer Expectation 

'favar(factor=3,horizon=48,rep=1000,ci=0.9,vd=1,save=1) 13 xdata xslow xir tcode yx_name @ ffr


After running this code, fill the boxes. (Please see the attachment)
Attachments
favar2step.PNG
favar2step.PNG (14.26 KiB) Viewed 253689 times

ouerk
Posts: 17
Joined: Tue May 23, 2017 11:29 am

Re: FAVAR add-in

Postby ouerk » Wed Jul 05, 2017 12:06 pm

Hi,
i have this ERROR message when i run the program : MATRIX SIZE MISMATCH

please can you help me ?
THANK you !!

dakila
Posts: 479
Joined: Tue Nov 24, 2015 4:57 pm

Re: FAVAR add-in

Postby dakila » Wed Jul 05, 2017 6:24 pm

if you have missing value, change the sample size.


Return to “Add-in Support”

Who is online

Users browsing this forum: No registered users and 10 guests