6. Stata loop (foreach) combined with local and levelsof functionality

[@muddisneakers]

People like you make the world a better place! Thank you so much for this video. I've been struggling with a solution and your levelsof explanation was just what I needed to solve the problem. 😊

[@learndataanalyticswithus5437]

Thanks for your appreciation!! :)

[@sayakdasgupta5309]

Hi, I am facing troubles with the tab type command. STATA is showing "variable type not found" (r111). I think it'll work if you let us know the variables in your updated file. Maybe, my auto2_updated file does not consist of all the variables that you have in your dta. Thanks in advance.

[@margretkiizza7890]

Thank you for the video. I am trying to run cross validation with kfold. However, I want to use the first 13 observations to train the poisson model and then validate it with the last 5 observations. and then make the predictions. However my do file is giving me errors. This is my dofile: local num_folds = 3
forval fold = 1/'num_folds' {
if `fold' 1 qui keep in 1/13
else if `fold' 2 qui keep in 1/13
else if `fold' 3 qui keep in 14/18
poisson NumberofAccidents Entrywidthe Entryangle Exitwidthex ExitRadius Cyclelanewidth InscribedDiameterD InAADTMajV lnAADTMajP lnAADTMinV lnAADTMinP, exposure(exposure)
predict estimatedaccidents if `fold' 3
list estimatedaccidents if e(sample), noobs}

[@learndataanalyticswithus5437]

As I don't have the data so I can't test your do file. However, the thing that can be a possible cause of error is that you are trying to run mode on very few observations.

[@shilpakrishnan7674]

could you explain a little bit on global macros in stata? how to write data path using global and to include global path while writing the path of dictionary file in stata?

[@learndataanalyticswithus5437]

Thanks Shilpa for your reply.
The given below code is a good example of how file paths can be used through globals.
// setting up globals
global projdir "C:/Users/amolr/Desktop/session_2"
global raw_data "${projdir}/01_data/01_raw_data"
global final_data "${projdir}/01_data/03_final_data"
// importing the vaccination dataset as of 2021.04.18
use "${raw_data}/covid_vaccination_2021_04_18.dta", clear

[@shilpakrishnan7674]

THANK YOU

[@learndataanalyticswithus5437]

You are welcome!

[@shilpakrishnan7674]

how to add the command replace variable x if y== some value within a loop? so that both x and y gets added by 1 in each iteration?

[@learndataanalyticswithus5437]

Can you explain your question with a numeric example? Sorry I am a bit unclear about your question. Thanks

[@shilpakrishnan7674]

@@learndataanalyticswithus5437 I think I need to use a nested loop. IDK how. The code without loop may look like this. How can I write a loop for this? Var_y is defined for Var_x>=5.
gen var_z =.
replace var_z=1 if Var_x ==1
replace var_z=1 if Var_x ==2
replace var_z=2 if Var_x ==3
replace var_z=3 if Var_x ==4
replace var_z=4 if Var_x == 5 & Var_y ==1
replace var_z=4 if Var_x == 5 & Var_y ==2
replace var_z=5 if Var_x == 6 & Var_y ==1
replace var_z=5 if Var_x == 6 & Var_y ==2

[@learndataanalyticswithus5437]

Maybe
replace 'i' if var_x>=5
In above eq i will be contacting variable y

[@varunmiglani110]

i want to get mean of two variable i run following command and i get error invalid syntax r(198);

foreach c inagcro ainc v_dairyinc {
tabstat `c'
}

[@learndataanalyticswithus5437]

What's the command you are running?

[@varunmiglani110]

can anyone help what error am i making

[@varunmiglani110]

foreach c inagcro ainc v_dairyinc {
tabstat `c'
}

[@learndataanalyticswithus5437]

Try foreach c of ...
Or try
foreach c in ....

[@varunmiglani110]

foreach c in .... worked