11:03 a.m.
Hi,
As in the documentation, I am trying to import data to gretl from a csv file.
The data is monthly, so it looks something like this:
,shillerRet,state,consSentiment,PMI
1987:02,0.0045773162,1,90.4,54.9
1987:03,0.004080661,1,90.2,52.6
1987:04,0.0045113495,1,90.8,55
1987:05,0.0050701111,1,92.8,55.5
1987:06,0.004516603,1,91.1,57.2
(...)
but gretl fails to interpret is as timeseries data:
parsing /home/ga1009/PhD/cpp/pmi/gretlInput.csv...
using delimiter ','
longest line: 43 characters
number of columns = 5
number of variables: 4
number of non-blank lines: 297
scanning for variable names...
line: ,shillerRet,state,consSentiment,PMI
scanning for row labels and data...
the cell for variable 3, obs 294 is empty: treating as missing value
the cell for variable 3, obs 295 is empty: treating as missing value
the cell for variable 3, obs 296 is empty: treating as missing value
first row label "1987:02", last label "2011:09"
trying to parse row labels as dates...
1987: probably a year... month 02?
1987:01: not a consistent date
but the dates are not complete and consistent
treating these as undated data
Am I doing something wrong?
Regards,
Grzegorz
3:34 p.m.
On Thu, 3 Jan 2013, Grześ Andruszkiewicz wrote:
As in the documentation, I am trying to import data to gretl from a
csv file.
The data is monthly, so it looks something like this:
,shillerRet,state,consSentiment,PMI
1987:02,0.0045773162,1,90.4,54.9
1987:03,0.004080661,1,90.2,52.6
1987:04,0.0045113495,1,90.8,55
1987:05,0.0050701111,1,92.8,55.5
1987:06,0.004516603,1,91.1,57.2
(...)
but gretl fails to interpret is as timeseries data [...]
I copied and pasted the sample above, and current gretl interpreted
it as time series OK (monthly, 1987:02 to 1987:06). Two questions
therefore suggest themselves:
(1) What version of gretl are you using? (Maybe the problem is due
to a bug that is already fixed.)
(2) What does the rest of the CSV file look like? (Maybe there's
some inconsistency in the remainder of the file that rules out a
consistent time series interpretation of the data.)
Allin Cottrell
3:35 p.m.
Hi,
I think that the name of the Date field is missing (there are some
rules for Gretl recognize CSV as time series).
Good luck.
Helio
Date,shillerRet,state,consSentiment,PMI
On Thu, Jan 3, 2013 at 4:03 PM, Grześ Andruszkiewicz <gandrusz(a)gmail.com> wrote:
Hi,
As in the documentation, I am trying to import data to gretl from a csv
file.
The data is monthly, so it looks something like this:
,shillerRet,state,consSentiment,PMI
1987:02,0.0045773162,1,90.4,54.9
1987:03,0.004080661,1,90.2,52.6
1987:04,0.0045113495,1,90.8,55
1987:05,0.0050701111,1,92.8,55.5
1987:06,0.004516603,1,91.1,57.2
(...)
but gretl fails to interpret is as timeseries data:
parsing /home/ga1009/PhD/cpp/pmi/gretlInput.csv...
using delimiter ','
longest line: 43 characters
number of columns = 5
number of variables: 4
number of non-blank lines: 297
scanning for variable names...
line: ,shillerRet,state,consSentiment,PMI
scanning for row labels and data...
the cell for variable 3, obs 294 is empty: treating as missing value
the cell for variable 3, obs 295 is empty: treating as missing value
the cell for variable 3, obs 296 is empty: treating as missing value
first row label "1987:02", last label "2011:09"
trying to parse row labels as dates...
1987: probably a year... month 02?
1987:01: not a consistent date
but the dates are not complete and consistent
treating these as undated data
Am I doing something wrong?
Regards,
Grzegorz
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
3:41 p.m.
On Thu, 3 Jan 2013, Hélio Guilherme wrote:
I think that the name of the Date field is missing (there are some
rules for Gretl recognize CSV as time series).
Actually, that should be OK. Our specification accepts an empty
first field on the first line (along with "obs", "date" and a few
other variants), as indicating a left-most column of observation
labels or dates.
Allin Cottrell
On Thu, Jan 3, 2013 at 4:03 PM, Grześ Andruszkiewicz
<gandrusz(a)gmail.com> wrote:
> Hi,
>
> As in the documentation, I am trying to import data to gretl from a csv
> file.
> The data is monthly, so it looks something like this:
> ,shillerRet,state,consSentiment,PMI
> 1987:02,0.0045773162,1,90.4,54.9
> 1987:03,0.004080661,1,90.2,52.6
6:54 p.m.
Thank you prof. Allin,
Sorry for not having taken the care of testing or reading the documentation.
Glad that you are always here to correct me :).
---
Now, looking for the import report, it seems that there is a 1987:01
row breaking the date series:
...
first row label "1987:02", last label "2011:09"
trying to parse row labels as dates...
1987: probably a year... month 02?
1987:01: not a consistent date
but the dates are not complete and consistent
treating these as undated data
...
My guess is a typo and the value should be 1987:10. Otherwise that row
should be reordered to be the first data row.
---
Helio
On Thu, Jan 3, 2013 at 8:41 PM, Allin Cottrell <cottrell(a)wfu.edu> wrote:
On Thu, 3 Jan 2013, Hélio Guilherme wrote:
> I think that the name of the Date field is missing (there are some
> rules for Gretl recognize CSV as time series).
Actually, that should be OK. Our specification accepts an empty first field
on the first line (along with "obs", "date" and a few other
variants), as
indicating a left-most column of observation labels or dates.
Allin Cottrell
> On Thu, Jan 3, 2013 at 4:03 PM, Grześ Andruszkiewicz <gandrusz(a)gmail.com>
> wrote:
>>
>> Hi,
>>
>> As in the documentation, I am trying to import data to gretl from a csv
>> file.
>> The data is monthly, so it looks something like this:
>> ,shillerRet,state,consSentiment,PMI
>> 1987:02,0.0045773162,1,90.4,54.9
>> 1987:03,0.004080661,1,90.2,52.6
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
4294
days inactive
4294
days old
4 comments
3 participants
participants (3)
-
Allin Cottrell -
Grześ Andruszkiewicz -
Hélio Guilherme