How can I create "Country Dummy" automatically?

[Myo Win]

Dear all,
Currently I am using 178 countries in my data set and I would like to use country fixed effect.

Now, dummy creation for each countries is as follow;

gen DAFG=0
replace DAFG=1 if iso3i=="AFG" | iso3j=="AFG"
gen DAGO=0
replace DAGO=1 if iso3i=="AGO" | iso3j=="AGO"
..................................................

.................................................. ...........

.................................................. ................
gen DVNM=0
replace DVNM=1 if iso3i=="VNM" | iso3j=="VNM"
gen DYEM=0
replace DYEM=1 if iso3i=="YEM" | iso3j=="YEM"
gen DZAF=0
replace DZAF=1 if iso3i=="ZAF" | iso3j=="ZAF"
gen DZMB=0
replace DZMB=1 if iso3i=="ZMB" | iso3j=="ZMB"
gen DZWE=0
replace DZWE=1 if iso3i=="ZWE" | iso3j=="ZWE"

The above command are manual command and please let me know automatically command to create 178 countries dummy.

Thank you very much.
Myo

[Marcos Almeida]

Welcome to the Stata Forum / Statalist,

Please read the FAQ. There you’ll find information about sharing data/command/output.

You may create these binary variables just by typing;

tabulate country, gen(mydummy)

That being said, maybe you don’t even need to do it.

[Jorrit Gosens]

Creating 178 dummy variables is probably not what you want.
Try for example, the below, with an option to do a fixed effect regression through xtreg, or by including dummies for all you countries with regress.
This does require that your country id variable is numeric; see help file of group for this purpose

webuse grunfeld
reg invest mvalue i.company
xtreg invest mvalue, fe

[Carlo Lazzaro]

Myo:
I do share previous comments.
As an aside, if you want to go -xt-, please note that the (highly) recommended layout is -long- (not -wide-, that is the one that your dataset in all likelihood adopts).
At the top of that, 178 dummies in -wide- format can hit (and probably overcome) -maxvar- limits and eat too many degrees of freedom.
See -help reshape- to change the layout of your dataset from -wide- to -long-..

[Myo Win]

Thank you very much for suggestions and all comments.

Thank you very much for your code @ Marcos. I can use like that;

tabulate iso3i, gen (A)

and then
A1 to A173 variables appear, and then I have to continue
tabulate iso3j, gen(B)
B1 to B173 variables define after above command.

I would like to create
C1= A1 & B1
C2= A2 & B2
....................................
....................................
...................................
C171= A171 & B171
C172= A172 & B172
C173= A173 & B173

How should I create C1 to C173 because I have two columns for countries, iso3i and iso3j.

[Marcos Almeida]

As I said, I think that: a) you probably don't need to create so many binary variables; b) to create these variables, just a single line of command would do (as I posted); c) Anyway, I fail to see the point of doing so. Perhaps you could explain why you should do it.

To end, maybe you just need to use the country variable as a cluster in the regression analysis. Hopefully that helps.

[Jorrit Gosens]

Why do you currently have two country name variables? Do they contain different info, or do the two lists have missing observations sometimes?
If it is the latter, do e.g.;

gen countrycode=iso3i
replace countrycode=iso3j if iso3i==""

If this is not what you want, explain the reason why you have two country code variables and what the result is that you want.

[Myo Win]

Dear Jorrit,
Thank you very much for your code and command. I would to use trade data and iso3i is export from country(I) to country (j) and iso3j is export from country j to country I. I want to measure FTA impact trade flow and want to use country fixed effect. Please check the following variables. I also agree creating 173 dummy variables is overcome. When I search the previous post I saw the following link;
https://www.statalist.org/forums/for...-fixed-effects
gen DAFG=0
replace DAFG=1 if iso3i=="AFG" | iso3j=="AFG"
gen DAGO=0
replace DAGO=1 if iso3i=="AGO" | iso3j=="AGO"
The above command are manual command and it can be wrong calculation. Please check iso3i and iso3j data as following;

iso3i	iso3j	DAFG	DAGO	DALB				DVNE	DZWE
EST	AFG	1	0	0				0	0
PAN	AFG	1	0	0				0	0
TJK	AFG	1	0	0				0	0
LSO	AFG	1	0	0				0	0
IRL	AFG	1	0	0				0	0
FIN	AFG	1	0	0				0	0
PHL	AFG	1	0	0				0	0
PRK	AFG	1	0	0				0	0
LBR	AFG	1	0	0				0	0
BEL	AFG	1	0	0				0	0
ALB	AFG	1	0	0				0	0
QAT	AFG	1	0	0				0	0
MYS	AFG	1	0	0				0	0
USA	AFG	1	0	0				0	0
DEU	AFG	1	0	0				0	0
SEN	AFG	1	0	0				0	0
LBY	AFG	1	0	0				0	0
ECU	AFG	1	0	0				0	0
SUR	AFG	1	0	0				0	0
ZWE	AFG	1	0	0				0	0
GNB	AFG	1	0	0				0	0
VCT	AFG	1	0	0				0	0
NER	AFG	1	0	0				0	0
BRN	AFG	1	0	0				0	0
AUS	AGO	0	1	0				0	0
EST	AGO	0	1	0				0	0
BWA	AGO	0	1	0				0	0
GNQ	AGO	0	1	0				0	0
DNK	AGO	0	1	0				0	0
MNE	AGO	0	1	0				0	0
ARG	AGO	0	1	0				0	0
HKG	AGO	0	1	0				0	0
BDI	AGO	0	1	0				0	0
SGP	AGO	0	1	0				0	0
LBY	AGO	0	1	0				0	0
KWT	AGO	0	1	0				0	0
MMR	AGO	0	1	0				0	0
TZA	AGO	0	1	0				0	0
CHE	AGO	0	1	0				0	0
THA	AGO	0	1	0				0	0
GTM	AGO	0	1	0				0	0
GRD	AGO	0	1	0				0	0
MWI	AGO	0	1	0				0	0
JAM	AGO	0	1	0				0	0
BEN	AGO	0	1	0				0	0
URY	AGO	0	1	0				0	0
AFG	AGO	1	1	0				0	0
ISL	AGO	0	1	0				0	0
UGA	AGO	0	1	0				0	0
BGD	AGO	0	1	0				0	0
YEM	AGO	0	1	0				0	0
BLR	AGO	0	1	0				0	0
JOR	AGO	0	1	0				0	0
LKA	AGO	0	1	0				0	0
UZB	AGO	0	1	0				0	0
HRV	AGO	0	1	0				0	0
DZA	AGO	0	1	0				0	0
MOZ	AGO	0	1	0				0	0
MDG	AGO	0	1	0				0	0
SWE	AGO	0	1	0				0	0
SVN	AGO	0	1	0				0	0
MEX	AGO	0	1	0				0	0
BEL	AGO	0	1	0				0	0
TLS	AGO	0	1	0				0	0
BLZ	AGO	0	1	0				0	0
KEN	AGO	0	1	0				0	0
ZAF	AGO	0	1	0				0	0
GIN	AGO	0	1	0				0	0
DEU	AGO	0	1	0				0	0
COG	AGO	0	0	1				0	0
ATG	AGO	0	0	1				0	0
MDA	AGO	0	0	1				0	0
ALB	AGO	0	0	1				0	0
PER	AGO	0	0	1				0	0
SAU	AGO	0	0	1				0	0
AGO	AGO	0	0	1				0	0
CIV	AGO	0	0	1				0	0
AZE	AGO	0	0	1				0	0
TUN	AGO	0	0	0				1	0
LVA	AGO	0	0	0				1	0
GMB	AGO	0	0	0				1	0
LCA	AGO	0	0	0				1	0
CYP	AGO	0	0	0				1	0
NIC	AGO	0	0	0				1	0
BGR	AGO	0	0	0				1	0
MUS	AGO	0	0	0				0	1
NPL	AGO	0	0	1				0	1
NAM	AGO	0	0	1				0	1
VNM	AGO	0	0	1				0	1
EST	AGO	0	0	1				0	1
SWZ	AGO	0	0	1				0	1
SLV	AGO	0	0	1				0	1
ROU	AGO	0	0	1				0	1

[Myo Win]

I am using following command for country fixed effect;
reg N_TF TA $xlist DAGO-DZWE if year==1990, robust
reg N_TF TA $xlist DAGO-DZWE if year==1995, robust
reg N_TF TA $xlist DAGO-DZWE if year==2000 robust


DAGO to DZWE will be 173 dummy variables. Actually i don't want to use above command for country fixed effect.

[Marcos Almeida]

If you don’t want to use the cornucopia of dummies, and considering the advice from this forum was against doing so, we reached an agreement. However, if you just want a friendlier command to create the dummies, fine, it was already given. This notwithstanding, the humoungous output seems to be something impossible to get rid of, considering the sort of analysis you chose.