Problem with loop r(198) invalid name

[Jaantje de Bruin]

Dear Statalist community,

I am using a loop with while in Stata 13.0 on Windows 10. I have the following code that I am running:

preserve
set seed 339487731
drawnorm SN_b1-SN_b26, n(1000000) means(e(b)) cov(e(V)) clear

postutil clear
postfile mypost prob_hat0 lo0 prob_hat1 lo1 hi1 diff_hat diff_lo diff_hi using "C:\Users\Anja\Documents\Esmées werk\Datasets\interaction_dataset2_test.dta", replace
noisily display "start"



local a = 0 
while ‘a’ <= 810 {
    {
scalar h_treatiesmem_open_a=5.686188
scalar h_gmi_a=238.3265 
scalar h_treatiesmem_open_b=5.736796 
scalar h_gmi_b=231.458  
scalar h_interaction_b=6544.354 
scalar h_contig=.3270718 
scalar h_comlang_ethno=.3071823 
scalar h_colony=.0729282 
scalar h_lndistcap=7.607629 
scalar h_lngdp1_a=8.343579 
scalar h_lngdp1_b=8.350814
scalar h_lnpop_a=16.56221 
scalar h_lnpop_b=16.48816 
scalar h_Landlocked_a=.1403315 
scalar h_Landlocked_b=.1546961     
scalar h_Agri_gdp_a=15.7774 
scalar h_Agri_gdp_b=17.01067 
scalar h_com_law_a=.2981366             
scalar h_com_law_b=.3240506 
scalar h_polity2_a=-.1072913     
scalar h_polity2_b=.2903407        
scalar h_constant=1                

generate x_betahat0 = SN_b1*h_treatiesmem_open_a ///
                        + SN_b2*‘a’ ///
                        + SN_b3*h_treatiesmem_open_a*‘a’ ///
                        + SN_b4*h_treatiesmem_open_b ///
                        + SN_b5*h_gmi_b ///
                        + SN_b6*h_interaction_b ///
                        + SN_b7*h_contig ///
                        + SN_b8*h_comlang_ethno ///
                        + SN_b9*h_colony ///
                        + SN_b10*h_lndistcap ///
                        + SN_b11*h_lngdp1_a ///
                        + SN_b12*h_lngdp1_b ///
                        + SN_b13*h_lnpop_a ///
                        + SN_b14*h_lnpop_b ///
                        + SN_b15*h_Landlocked_a ///
                        + SN_b16*h_Landlocked_b ///
                        + SN_b17*h_Agri_gdp_a ///
                        + SN_b18*h_Agri_gdp_b ///
                        + SN_b19*h_com_law_a ///
                        + SN_b20*h_com_law_b ///
                        + SN_b21*h_polity2_a ///
                        + SN_b22*h_polity2_b ///
                        + SN_b23*h_constant 
                        
generate x_betahat1 = SN_b1*(h_treatiesmem_open_a+1) ///
                        + SN_b2*‘a’ ///
                        + SN_b3*(h_treatiesmem_open_a+1)*‘a’ ///
                        + SN_b4*h_treatiesmem_open_b ///
                        + SN_b5*h_gmi_b ///
                        + SN_b6*h_interaction_b ///
                        + SN_b7*h_contig ///
                        + SN_b8*h_comlang_ethno ///
                        + SN_b9*h_colony ///
                        + SN_b10*h_lndistcap ///
                        + SN_b11*h_lngdp1_a ///
                        + SN_b12*h_lngdp1_b ///
                        + SN_b13*h_lnpop_a ///
                        + SN_b14*h_lnpop_b ///
                        + SN_b15*h_Landlocked_a ///
                        + SN_b16*h_Landlocked_b ///
                        + SN_b17*h_Agri_gdp_a ///
                        + SN_b18*h_Agri_gdp_b ///
                        + SN_b19*h_com_law_a ///
                        + SN_b20*h_com_law_b ///
                        + SN_b21*h_polity2_a ///
                        + SN_b22*h_polity2_b ///
                        + SN_b23*h_constant 

gen prob0=logistic(x_betahat0)
gen prob1=logistic(x_betahat1)
gen diff=prob1-prob0 

egen probhat0=mean(prob0)
egen probhat1=mean(prob1)
egen diffhat=mean(diff)

tempname prob_hat0 lo0 hi0 prob_hat1 lo1 hi1 diff_hat diff_lo diff_hi

_pctile prob0, p(2.5,97.5) 
scalar ‘lo0’ = r(r1)
scalar ‘hi0’ = r(r2)

_pctile prob1, p(2.5,97.5)
scalar ‘lo1’= r(r1)
scalar ‘hi1’= r(r2)

_pctile diff, p(2.5,97.5)
scalar ‘diff_lo’= r(r1)
scalar ‘diff_hi’= r(r2)

scalar ‘prob_hat0’=probhat0
scalar ‘prob_hat1’=probhat1
scalar ‘diff_hat’=diffhat

post mypost (‘prob_hat0’) (‘lo0’) (‘hi0’) (‘prob_hat1’) (‘lo1’) (‘hi1’) ///
(‘diff_hat’) (‘diff_lo’) (‘diff_hi’)

    }

drop x_betahat0 x_betahat1 prob0 prob1 diff probhat0 probhat1 diffhat
local a=‘a’+ 1
display "." _c

}


display ""
postclose mypost

restore
merge using "C:\Users\Anja\Documents\Esmées werk\Datasets\interaction_dataset2_test.dta"

After executing, Stata is showing me the error (198): 'a' invalid name. When I try to trace the error it seems that Stata does not recognize 'a' the second time it runs the loop.

I have tried to change '' into ‘’, I've run my entire code (and not only the loop), but nothing seems to solve the issue.

Some more information: I use an ordered logit model and the dataset contains 23 variables. I use the loop to make a plot that shows the interaction between treatiesmem_open_a and gmi_a (following Matt Golder's example from http://mattgolder.com/files/interact...teraction3.pdf). The variable of interest ranges from 0-810 (that why I use the command while.) After scalar the mean of each variable is given.

Thank you in advance. I am happy to provide more information.

Jaantje

[Ken Chui]

Welcome to Statalist.

I'm not sure if that's the problem, but the grammatical rule to refer to a local/global variable is this:

`the_var'

They are asymmetric quotes. The opening one is "grave accent", in US keyboard it's at the top left, next to "1" and often keyed together with tilde (~). The closing one is the usual single quotation mark.

Also, make sure you're using straight quotation mark. If you typed all the code inside Stata do file editor or any code editor, it should be fine. If you have edited the code in MS Word, etc. The quote may be a slanted formatted quote, and I believe in earlier Stata version they may not work.

[Jaantje de Bruin]

Thank you so much for your fast reply! Your suggestion worked.

Best,
Jaantje

[Ken Chui]

Glad that helped. Good luck with the analysis.

[Jaantje de Bruin]

Thank you!