Dealing with thirteen periods per year

[ech447]

I am a long-standing user of EViews. I am now dealing with a data set that has thirteen periods per year. Nothing exotic; that's just how the data was collected. Thus far, I have treated it as an irregular time series, so EViews deals with it as having just thirteen days per year. This doesn't seem to create problems (though some funny looking graphs), and I can use regression, VAR's, Causality tests, correlograms, etc. All the good stuff.
I have two questions.
1. Is there something here that I am missing, so computations may not be going as they should? (I have checked a few with spreadsheet calculations, and they have all checked. But, it would be extremely tedious to check them all.)
2. Is there a more natural way to get EViews to do this? Its just like monthly data, except with thirteen months.

[EViews Glenn]

For the irregular data, EViews treats lags as between the sequential observations so you shouldn't have any problems with analysis. As to the funny looking graphs, what are you using as your irregular data identifier? What's funny about what you are seeing?

[ech447]

Thanks.
The line graphs are sequences of dots instead of line graphs. I will play with that and see if there is a "connect the dots" command that works for irregular data. (I forgot to mention my version in my question--EViews 7.1 updated as of yesterday.)
Thanks again.

[EViews Glenn]

There should be a "sample break and NA handling" option that you can set to "connect adjacent" when you are displaying your line graphs.

[ech447]

Thanks.

[jalle7]

Hi,

How do I set up irregular dated time series data on eviews? It says that irregular dated workfiles may be made from unstructured workfiles by later changing the date, how do I do this?

Does using irregular dated time series data mean that I cannot lag my variables?

Thanks

[EViews Gareth]

Create an undated first. Then create a series in the workfile that contains the dates you want. Then restructure the page to be irregular dated using that date series.

Yes, you can still use lags.