Trouble changing sample in a script
by Logan Kelly
Hello,
I am having trouble changing the sample period in a script file. I initialy set the sample using
smpl 1967:1 2007:4
Then latter in the script I set the sample to
smpl 1967:1 2011:4
But when I estimate a VAR the sample period remains 1967:1 - 2007:4
I have also used the command
smpl --full
with the same results. The sample period does not change for the estimation.
Two questions:
Can the smpl command be used in a loop?
What is the correct way to change the sample period in a script?
12 years, 11 months
Import data as panel or change existing data set structure
by Jan Tille
Hello dear gretl users,
I have a question regarding the structuring of data as panel-set.
I want to create a panel that consists of 890 investment fund time series. I have no problem to stack them either cross-sectionally or as a stacked time-series.
The problem I am trying to solve is that I also have a single time series that shall be used as common a regressor for all the individual units, i.e. I need this series in a panel setup.
I know that I can prepare an excel sheet, where I either copy the time-series 890 times in a single column (stacked time-series) or where I copy each time observation 890 times (cross-section). As this is a pretty dull task I am looking for a command so that gretl performs this task upon the data import.
I hope, that I have presented my problem properly to you.
Many thanks in advance,
Jan
12 years, 11 months
local quantile regression
by Erman Terciyanlı
Hi,
I am working on quantile regreession for estimating different quantiles of our forecast in gretl for a while. Actually we have succeed to call quantreg from our C code but we additionally need to make it for local quantiles that have weights. Hence we need to estimate quantiles like wls model using weights between local forecast and observations. Is it possible in gretl or better to use another tool?
Regards,
Erman
12 years, 11 months
OLS problem
by Emiel Denuwelaere
Hello,
For my school these, i am doing research to investigate the effect of
different variables on the reserve currency shares of different currencies.
Before i am working with panel data, i liked to make an OLS estimation on
the dollar. So i'm using time series data. I checked the variables for
stationarity using the ADF-test, and it seems that the dependent variable
is non-stationary along with two explanatory variables. The other 4
explanatory variables are stationary. Since you cannot run regressions with
variables of a different integration, i am not sure how to proceed.. I
suggest it is not appropriate to just take the first differences of the
non-stationary variables?
Any help on this topic would be very much appreciated. Thanks in advance.
For additional info on the data/research, please do not hesitate to contact
me.
--
Met vriendelijke groeten,
Emiel Denuwelaere
M: +32 (0)476 56 08 75
E: emiel.denuwelaere(a)gmail.com
12 years, 11 months
problem (or bug?) with loop and strings
by artur tarassow
Hey all,
I've got a problem with the following example. I want to run loop "j" over
different samples where the starting date and ending date are given by
strings (from @temp to @last). If you run this loop "j" separately it
works, but embedding it into the loop "i" does give the error message:
"Data types not conformable for operation".
<hansl>
open australia
sprintf first "%s", obslabel($t1)
sprintf temp "%s", obslabel($t1+2)
sprintf last "%s", obslabel($t2)
printf "Full sample: @first to @last.\n"
scalar rep = 10
loop i=1..rep
smpl @first @temp
loop j=@temp..@last
scalar x=0
smpl ; +1
x=x+1
endloop
endloop
<\hansl>
I cannot figure out the problem here.
Best,
Artur
12 years, 11 months
Size of Diagram
by Philipp Wehling
Hi there,
I am using gretl for making statistical analysis for my master thesis. I
have to calculate many diagrams (x-y-time-graph). I also have to set my
focus on different parts of my graph.
It is possible to zoom in the parts. But it would be great to see a
larger view of the whole diagram. I think the window is very small and I
cant maximize it.
Did I miss some options or features?
Best Regards,
Philipp
12 years, 11 months
