Summarizing essential information by some conditions

[Maria Ivanova]

Hello!
Now I am working with many files from 2006 to 2017 (each file for each year). In every of them there are date and month of interview, cost for food in this period, year of birth of a person and so on. I need to make a column with the number of week (1, 2, 3, 4) for each month and a column with money that he/she spent on food. I have already made a column for weeks. So, for every row I defined a number from 1 to 4. I do not know how to calculate the sum of costs for a particular week and month. Eventually, I need to make a linear graph, where on X axis I will have the number of week and month, Y axis - costs.
For example,
week month costs
4 12 200
1 11 500
2 10 300
3 10 550
4 12 350
1 11 400
2 12 250
And I think that I need to get something like
week month costs
1 11 900
2 10 300
2 12 250
3 10 550
4 12 550
Later I suppose I will need to sort it in order to make a graph, but, actually, now I do not know, how to draw it, when I have two conditions on a week and a month.
Could you help me with these problems, please?
Thank you in advance

[Andrew Musau]

Here is a sketch. You will need to incorporate year into the code given that you have multi-year data. With many time periods, you may also consider switching the axes to improve readability. I use labmask from Stata Journal, authored by Nick Cox.

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(week month costs)
4 12 200
1 11 500
2 10 300
3 10 550
4 12 350
1 11 400
2 12 250
end

bys week month: egen tc= total(costs)
egen time= group(week month)
label define month 1 "Jan" 2 "Feb" 3 "Mar" 4  "Apr" 5 "May" 6 "Jun" 7 "Jul" 8 "Aug" 9 "Sep" 10 "Oct" 11 "Nov" 12 "Dec"
lab values month month
decode month, gen(strmonth)
gen lab_time= "Week "+ string(week) + " " +strmonth
labmask time, values(lab_time)
line tc time, xlab(, val) ytitle("Expenditure") xtitle("") xscale(r(0.75 5.3)) scheme(s1color)

ADDED IN EDIT: Note that the x-axis is not correctly ordered since I grouped by week-month and not month-week. So you want to group by year-month-week. The best way to achieve this is to create a time variable that Stata understands.

[Maria Ivanova]

Here is a sketch. You will need to incorporate year into the code given that you have multi-year data. With many time periods, you may also consider switching the axes to improve readability. I use labmask from Stata Journal, authored by Nick Cox.

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(week month costs)
4 12 200
1 11 500
2 10 300
3 10 550
4 12 350
1 11 400
2 12 250
end

bys week month: egen tc= total(costs)
egen time= group(week month)
label define month 1 "Jan" 2 "Feb" 3 "Mar" 4 "Apr" 5 "May" 6 "Jun" 7 "Jul" 8 "Aug" 9 "Sep" 10 "Oct" 11 "Nov" 12 "Dec"
lab values month month
decode month, gen(strmonth)
gen lab_time= "Week "+ string(week) + " " +strmonth
labmask time, values(lab_time)
line tc time, xlab(, val) ytitle("Expenditure") xtitle("") xscale(r(0.75 5.3)) scheme(s1color)

[ATTACH=CONFIG]n1541240[/ATTACH]




ADDED IN EDIT: Note that the x-axis is not correctly ordered since I grouped by week-month and not month-week. So you want to group by year-month-week. The best way to achieve this is to create a time variable that Stata understands.

Andrew, thank you very much!
I will try.