I'ma beginner in Stata and I have some questions on observing the jumps of number of observations and dropping the outliers.
So my questions are as follows:
1)
For instance, I have a firm-level panel data which looks like:
Year | firm number | sector |
1993 | 1 | Manufacturing |
1994 | 1 | Manufacturing |
1993 | 2 | Sevice |
1994 | 2 | Service |
1995 | 2 | Service |
1994 | 3 | Transport |
1995 | 3 | Transport |
sector | 1993 | 1994 | 1995 |
Manufacturing | 1 | 1 | . |
Service | 1 | 1 | 1 |
Transport | . | 1 | 1 |
I would like to know the command for reshaping this data..
2)
If I say that there is a jump: (for instance) for manufacturing sector, in 1993 there are 103 firms, but in 1994, there are 902 firms in total.
In order to observe this, how can I count this firms in order to see the jumps ?
3)
In addition, how can I drop the outliers that has a dramatic jump in my dataset (as manufacturing from my example)?
Thank you very much,
Anne-Claire
0 Response to Observe the jumps of (number of) observations and drop the outliers
Post a Comment