So I'm now using the IES data2010/2011 cross-sectional data of South Africa. I'm not sure if i was right in using the data in stata.
So, first ,there are 10 broad categories expenditure in the dataset. I choosed 4 from them as a group system:agriculture, food, transport and metal-products.
first , i calculated the expenditure share of each category ,w1=agriculture expenditure/(agriculture+ food+transport metal-products expenditure)=agriculture expenditure/total_exp2 ,w2,w3,w4. So that w1+w2+w3+w4=1
there are price level of these four categories in the dataset. So i gen p1=price_expenditure ,p2,p3,p4
because in the statistics Journal ,it says the value of α0 should be slightly less than the lowest value of lnm observed in the data.So I loged the whole expenditure( total_exp2)if these four caterories, i got the value of 9.3 ,so that i choose 9 as the value of α0. the demographic, i choose the income docile which is "dec_inc" dec_inc range from 1 to 10
The command are as follows:
net sj 12-3 st0268
net install st0268
quaids w1-w4, anot(9) prices(p1-p4) expenditure( total_exp2 ) demographics( dec_inc ) nolog
estat expenditure e*
summarize e_1-e_4
estat uncompensated if dec_inc==1, atmeans
matrix x1 = r(uncompelas)
matrix list x1
estat uncompensated if dec_inc==2, atmeans
matrix x2 = r(uncompelas)
matrix list x2
..........
This is what i have done so far , but i'm really not sure if the whole process is correct or not .Hope somebody can correct my work ,thank you .
So, first ,there are 10 broad categories expenditure in the dataset. I choosed 4 from them as a group system:agriculture, food, transport and metal-products.
first , i calculated the expenditure share of each category ,w1=agriculture expenditure/(agriculture+ food+transport metal-products expenditure)=agriculture expenditure/total_exp2 ,w2,w3,w4. So that w1+w2+w3+w4=1
there are price level of these four categories in the dataset. So i gen p1=price_expenditure ,p2,p3,p4
because in the statistics Journal ,it says the value of α0 should be slightly less than the lowest value of lnm observed in the data.So I loged the whole expenditure( total_exp2)if these four caterories, i got the value of 9.3 ,so that i choose 9 as the value of α0. the demographic, i choose the income docile which is "dec_inc" dec_inc range from 1 to 10
The command are as follows:
net sj 12-3 st0268
net install st0268
quaids w1-w4, anot(9) prices(p1-p4) expenditure( total_exp2 ) demographics( dec_inc ) nolog
estat expenditure e*
summarize e_1-e_4
estat uncompensated if dec_inc==1, atmeans
matrix x1 = r(uncompelas)
matrix list x1
estat uncompensated if dec_inc==2, atmeans
matrix x2 = r(uncompelas)
matrix list x2
..........
This is what i have done so far , but i'm really not sure if the whole process is correct or not .Hope somebody can correct my work ,thank you .
