Date axis

[Katjes]

Hi,
the application: I have calculated the autocorrelation function of two series for lag 1 to 28, saved them in a vector. However, because the graph editor for matrices is not the same as for time series (less possibilities), I converted the vectors back to series, grouped them.
the problem:
I'm not able to manipulate the date axis. So far I reached what I was looking for, but the space between the two digit numbers is too small, hence I would like to show only 1 3 5 .. or any other steps...
Maybe you further know a better way to change the axis to integer numbers instead of date??

Here my code:

Code: Select all

smpl 1996Q1 1996Q1+27 ' I reduced the sample size that the vector dimension fits to the sample
mtos(AR_EU_n, AAR_EU_n)
mtos(AR_EU_t, AAR_EU_t)

group AAR_EU AAR_EU_n AAR_EU_t

freeze(fig_AR_EU) AAR_EU.line

fig_AR_EU.axis(l) font("Arial", 12i) zeroline  'adds the zeroline
fig_AR_EU.axis(b)  font("Arial", 12) angle(0)  'doesnot turn date

fig_AR_{%c}.setobslabel(series) ar_zeit ' AR_ZEIT is a alpha series with the values 1 to 28


Thanks for suggestions,
Katjes

[EViews Pamela]

Use the datelabel command with the interval option. For instance, to label every 5 observations:

fig_AR_EU.datelabel interval(obs, 5)

Pamela

[Katjes]

Thanks Pamela, it works perfectly and looks great!

[Gohou]

hope you are well. I am doing a research and Eviews beginner .. could someone please help me finding how to plot data as month instead of year on my graph?

Regards

[EViews Pamela]

Double click the graph to bring up the main graph dialog. In the Axes & Scaling -> Obs/Date axis page, there is a groupbox labeled Date format. Use the drop-down to select the date type you would like.

Pamela

[Gohou]

Thanks a lot Pamela.
Regards