Zero Expenditure Share with Quaids or Quaidsce

[Mahedi Hassan]

Hello Everyone,

I am trying to run Quaids model using Household Income and Expenditure survey. There is zero expenditure share in the household. If I run quaids model, observations with zero expenditure share are dropped. If I run the quaidsce model, 'noncensoring is found the dataset' errors is showing. What can be done in this regards? Please help me out of this. Thanks in advance.

Mahedi

[Henry Hu]

Hey, I'm having the same issue here, did you manage to find a solution?

If you use the option "nocensor" does it run at least?

One thing I can't seem to wrap my head around is this:
1.) Let's assume we need to aggregate our good items. So we have items 1, 2, and 3 in Category A; and items 4, 5 in Category B.
2.) If a household experiences 0 expenditure in items 1, 2, & 3 simultaneously, how would you calculate the price index for Category A for that household?

The shares are now 0 for all 3 goods in Category A, so the price index becomes 0 too. But the QUAIDS command in STATA complains that this is an issue (and obviously it is).