Panel data sampled at irregular time intervals

[airshu23]

Hi,

I have cancer mortalitity data from 1974, 1991, and 2000 for 15 chinese provinces. I am looking at the relationship between gdp per capita and cancer mortality (dependent variable).

How am I supposed to structure the workfile given irregular time inteverals. I think I am supposed to pick "irregular", but I am unsure how to proceed after that.

(I ran a test regression placing cancer mortality data in fictitious years 1998 (data from 1974) ,1999 (data from 1991), and 2000. Would that estimate give the same result?)

Thanks in advance for your help.

[EViews Gareth]

EViews should just automatically detect the panel nature. It won't care about the irregular times.

[airshu23]

Which option do I use to structure the workfile? (I have eviews 6 student version)

Balanced panel with start date of 1974 and end date of 2000, thereby creating observations which will not be filled?

[EViews Gareth]

Where is the data?