One line code for performing arithmetic operation (here division) for multiple variables

[Aishwary Trivedi]

Exchange rate CPI.dta Here is the dataset and I want a proportionate values for all countries divided by United States for all time periods. What will be a one line code for that? [For Ex: I need Afghanistan/U.S. (for all years), Australia/U.S. (for all years), Austria/U.S. (for all years), Azerbaijan/U.S. (for all years) and so on]

year	Afghanistan, Islamic Rep. of	Albania	Algeria	Angola	Anguilla	Antigua and Barbuda	Armenia, Rep. of	Aruba, Kingdom of the Netherlands	Australia	Austria	Azerbaijan, Rep. of
1989			13					47	56	64
1990			16	0	52			49	60	66
1991		6	20	0	54			52	62	68	0
1992		18	26	0	55			54	62	71	0
1993		33	31	0	58		0	57	63	73	0
1994		41	40	0	60		16	60	65	75	8
1995		44	52	0	60		44	62	68	77	43
1996		50	62	0	63		53	65	69	78	51
1997		66	65	0	63		60	66	70	80	53
1998		80	68	0	64	80	65	68	70	80	52
1999		80	70	1	65	80	66	69	71	81	48
2000		80	70	3	68	81	65	72	74	83	49
2001		83	73	7	72	82	67	74	78	85	50
2002		89	74	15	72	84	68	77	80	86	51
2003		90	78	30	74	86	71	79	82	87	52
2004	64	92	81	44	77	88	76	81	84	89	56
2005	72	94	82	54	81	89	76	84	86	91	61
2006	76	96	84	61	88	91	79	87	89	93	66
2007	83	99	87	68	92	92	82	92	91	95	77
2008	105	102	91	77	99	97	89	100	95	98	93
2009	98	104	96	87	98	97	92	98	97	98	95
2010	100	100	100	100	100	100	100	100	100	100	100
2011	112	103	105	113	105	103	108	104	103	103	108
2012	119	106	114	125	106	107	110	105	105	106	109
2013	128	108	118	136	106	108	117	102	108	108	112
2014	134	109	121	146	106	109	120	103	110	110	113
2015	133	113	127	159	105	110	125	103	112	111	118
2016	139	113	135	208	105	110	123	102	113	112	132
2017	146	115	142	271	106	112	124	101	116	114	150

[Øyvind Snilsberg]

ds year ,not
foreach v in `r(varlist)' {
g new`v' = `v' / UnitedStates
}

[Nick Cox]

This can probably be done in one line but you need to reshape long first.

Reading list:

https://www.statalist.org/forums/help#stata explains why .dta attachments are less helpful than using dataex. In your case. it is clear that most country names could not be Stata variable names.

Speaking Stata: Compared with … (sagepub.com) runs through some useful technique.

[Carlo Lazzaro]

Aishwary:
the following code does the trick, provided that you shorten the name of most of the nations:

order UnitedStates, last
foreach var of varlist AfghanistanIslamicRepof - Zimbabwe {
  2. gen `var'_ratio=`var'/UnitedStates
  3.  }

[Aishwary Trivedi]

ds year ,not
foreach v in `r(varlist)' {
g new`v' = `v' / UnitedStates
}

:Thanks so much it did work for me. Suppose, I have two datasets given below, now how can I merge the datasets so that resulting values in the merged dataset gets multiplied with each other. How should i go about it? For example: Afg (2004) = 48[1st dataset] * Afg (1989) = 64 [2nd dataset], Algeria (1989)= 8{1st dataset] * Algeria(1989) = 13[2nd dataset] and so on.

year	Afg	Albania	Algeria	Angola	Anguilla
1989	39		8	0	3
1990	39		9	0	3
1991	39		18	0	3
1992	39	75	22	0	3
1993	39	102	23	0	3
1994	39	95	35	0	3
1995	37	93	48	0	3
1996	48	104	55	0	3
1997	48	149	58	0	3
1998	48	151	59	0	3
1999	47	138	67	3	3
2000	47	144	75	10	3
2001	48	143	77	22	3
2002	47	140	80	44	3
2003	49	122	77	75	3
2004	48	103	72	84	3

	Afg	Albania	Algeria	Angola	Anguilla
1989			13
1990			16	0	52
1991		6	20	0	54
1992		18	26	0	55
1993		33	31	0	58
1994		41	40	0	60
1995		44	52	0	60
1996		50	62	0	63
1997		66	65	0	63
1998		80	68	0	64
1999		80	70	1	65
2000		80	70	3	68
2001		83	73	7	72
2002		89	74	15	72
2003		90	78	30	74
2004	64	92	81	44	77

[Carlo Lazzaro]

Aishwary:
1) you should -append- (not -merge-) your datasets first;
2) after that, the following toy-example can do the trick:

. foreach var of varlist Afg-Anguilla {
  2. bysort year: g wanted_`var'=  `var'[_n]* `var'[_n-1]
  3.  }


. list

     +-----------------------------------------------------------------------------------------------------------+
     | year   Afg   Albania   Algeria   Angola   Anguilla   wanted~g   want~nia   want~ria   want~ola   want~lla |
     |-----------------------------------------------------------------------------------------------------------|
  1. | 1989    39         .         8        0          3          .          .          .          .          . |
  2. | 1989     .         .         .       13          .          .          .          .          0          . |
  3. | 1990    39         .         9        0          3          .          .          .          .          . |
  4. | 1990    39         .        16        0         52       1521          .        144          0        156 |
     +-----------------------------------------------------------------------------------------------------------+

.