Hi,
I am currently investigating the effects of labour demand on house prices within the UK. I have data on yearly changes in house prices and labour demand for 322 English local authorities across 2016-2021.
I am nearing the end of my investigation and am trying to investigate the different effects of labour demand on house prices at different regions throughout the country.
I have grouped the 322 local authorities into 9 groups, which represent south east, south west, London etc.
I am currently using the following command:
xtreg house_price c.labour_demand##i.region control1 control2 i.year, fe cluster(ID)
I then observe a singular 'labour_demand' coefficient at the top, and then observe multiple interaction term coefficients below.
I have also been using c.labour_demand##ibN.region, where N is the group identification. I have used this to change which region is used at the reference region.
I was wondering if anyone could help me with the interpretation of the coefficients - is the 'labour_demand' coefficient (at the top) the effect of labour demand on house prices for the reference region, or the average effect of all regions? I had read that this coefficient should be the average effect of all regions, however the value changes significantly when I change the reference region - which makes me think it cannot be the average?
I am also under the impression that the coefficients on the interaction term below are the effect of labour demand on house prices in each region RELATIVE to the reference region, is this correct?
If anyone could help me with this it would be really appreciated!
Thanks.
I am currently investigating the effects of labour demand on house prices within the UK. I have data on yearly changes in house prices and labour demand for 322 English local authorities across 2016-2021.
I am nearing the end of my investigation and am trying to investigate the different effects of labour demand on house prices at different regions throughout the country.
I have grouped the 322 local authorities into 9 groups, which represent south east, south west, London etc.
I am currently using the following command:
xtreg house_price c.labour_demand##i.region control1 control2 i.year, fe cluster(ID)
I then observe a singular 'labour_demand' coefficient at the top, and then observe multiple interaction term coefficients below.
I have also been using c.labour_demand##ibN.region, where N is the group identification. I have used this to change which region is used at the reference region.
I was wondering if anyone could help me with the interpretation of the coefficients - is the 'labour_demand' coefficient (at the top) the effect of labour demand on house prices for the reference region, or the average effect of all regions? I had read that this coefficient should be the average effect of all regions, however the value changes significantly when I change the reference region - which makes me think it cannot be the average?
I am also under the impression that the coefficients on the interaction term below are the effect of labour demand on house prices in each region RELATIVE to the reference region, is this correct?
If anyone could help me with this it would be really appreciated!
Thanks.
Comment