creating means of neighbours' values without spatial coordinates (cross-section data)

[Afarebea Agyen]

Dear all,

I am trying to create a variable which is the mean of another variable of all observation but the current one in the same location. In the data I have the following variables; observation_id, location_id, and rate, which are all numeric. In the data below for example, I want to create n_rate(label as neighbours' rate) i.e. I want the average rate of neighbours assuming that observations with the same location id are neighbours.

Observation_id	Location_id	rate	n_rate
1	1	5	4.5
2	1	6	4
3	1	3	5.5
4	2	2	4.5
5	2	4	3.5
6	2	5	3
7	3	7	2.75
8	3	3	3.75
9	3	5
10	3	2
11	3	1
12	4	3
13	4	4
14	5	2
15	5	6
16	5	3
17	5	5
18	6	4
19	6	1

bys location_id: egen Neighbours_rate=mean(rate)

gives me the mean of all observations in each location but I want to exclude current observations in each case.

I would be grateful for assistance.
Thank you.