"Fan charts" and processing of publication lags

[PMaier]

Good morning,

I'm wondering whether others have found simple solutions to the following two issues.

First, has anyone figured out a simple way to generate fan charts in EViews (http://en.wikipedia.org/wiki/Fan_chart_ ... _series%29)? I didn't find anything in the forum.

Second, economic data is released sequentially. When I retrieve monthly data from Haver, I can get a "time stamp" of the day when the data was last updated. I'm trying to figure out a simple way to update a forecasting model sequentially, and print for each new piece of data (i) when it was last updated, and (ii) the marginal impact of that piece of information (plus, ideally, the "cumulative" impact of multiple pieces of information. Say, a table such as (just making up numbers):

Forecast as of March 1: US GDP 2.1%

Data release since March 1 Impact on U.S. GDP Cumulative impact
- ADP (March 2) +0.4 +0.4
- ISM (March 3) -0.3 +0.1
- Consumer sentiment (March 4) +0.5 +0.6

etc.

One way of doing this would be to have in the Excel data file a table with all indicators and the date they were last updated. I could read in that table and save this information for each series in EViews. Then, I could order the variables in the right sequence and process them in this order. A bit clumsy, but should work.

In practice, though, a set of complications arise. How to best sort the variables in the right order (in Excel? In Eviews?) To make it even more general, what's the best way to figure out when the model was last run (I could output a file with that information, I guess)?

In short, if anyone has attempted to crack these issues and has found a simple way, I'd be happy to hear it....

Many thanks,

Philipp

[EViews Glenn]

I'll answer the first quickly quickly and leave the other for the moment since it's a bit more involved and someone may be able to come up with a better solution than I.

You want a band graph in EViews. You'll create a group with two series representing the band and then a series representing the middle line.