how to convert character to numeric ?

[srikanthyadav44]

my dataset has some numeric and some character variables 

company_name	esg2021	esg2020	esg2019	esg2018
Talwalkars Better Value Fitness Ltd
Tarapur Transformers Ltd
GB Global Ltd
Jaypee Infratech Ltd
SJVN Ltd
Rainbow Papers Ltd
Sundaram Multi Pap Ltd
Pioneer Distilleries Ltd
Technofab Engineering Ltd
Denso India Pvt Ltd
Manappuram Finance Ltd	53.26	56.28
Lakshmi Finance and Industrial Corp Ltd
Hindustan Media Ventures Ltd
Tilaknagar Industries Ltd
Indian Metals and Ferro Alloys Ltd
Bliss GVS Pharma Ltd
Emami Realty Ltd
Bajaj Consumer Care Ltd
Prakash Steelage Ltd
Asian Hotels (West) Ltd
Sakthi Finance Ltd
Asian Hotels East Ltd
Industrial Investment Trust Ltd

the data is relating to esg of the companies from esg2021 to esg2001(ie.,year 2021 to 2001)

in this except, company_name, all other variables should be in numeric form. but, some of the esg variables are in character and some are in numeric form.

how to convert all the esg variables in to numeric  

thanks in advance 

[Patrick]

WHY are some of the esg... variables character? Are you importing from an Excel?

[Astounding]

SAS made those variables character for a reason. You need to examine the data first and decide how you want to handle the values that won't convert such as

NA
>99
<1

SAS didn't want to make those variables character. It was forced to do so because some of the values contain characters.

[andreas_lds]

---

@srikanthyadav44 wrote:

my dataset has some numeric and some character variables 

in this except, company_name, all other variables should be in numeric form. but, some of the esg variables are in character and some are in numeric form.

....

how to convert all the esg variables in to numeric  

---

By fixing the step reading the file. If the source is an excel-file, convert it to csv and write a data step to read it. Anything else will fail sooner or later.

[robsonandrade0 · Accepted solution]

you must put a calculated column like for example:

esg2021_new = esg2021 * 1;

so this new variable becomes numerical

[andreas_lds]

---

@robsonandrade0 wrote:

you must put a calculated column like for example:

esg2021_new = esg2021 * 1;

so this new variable becomes numerical

---

Bad idea, using this code notes (that should be at least warnings) are written to the log about automatic type conversation. Most likely causing errors if esg2021 has non-numeric content.

esg_2021_new = input(esg2012, ?? best.);

The ?? take care of values that aren't numeric, the input-function returns a missing value in those cases.

The data seems to be in spreadsheet-style, easy to read for humans, difficult to work with in code. So i would do something like this

/* untested code */
data transposed;
  set have;
  length year esg 8;
  array values[2001:2021] esg2001-esg2021;

  do year = lbound(values) to hbound(values);
    esg = input(values[year], ?? best.);
    
    if missing(esg) and not missing(values[year]) then do;
      /* maybe write a note to the log if conversion failed */
    end;

    output;
  end;
  drop i esg2001-esg2021;
run;

Maybe using proc transpose before converting the esg-variables is even better, but i am to lazy to change the code now 😉

[srikanthyadav44]

your technique to convert the character variable to numeric is working. 

can you please help me out in writing the SAS CODE  with array function for the same purpose, because i have large number of variables to convert them into numeric. 

i have some numeric and character variables in the file. i think it is better to multiple all the variables with 1 and create a new variable and them remove the old one. 

it can be done easy, if i use array function. please help me out in this regard. 

i have to do the following functions with the code

1. except date, i have to multiple all variables from Y1 to Y100  with 1 and create a new variable

2. remove the old variable

3. assign old variable name to new variable 

the model of my input file is as follows.

Date	Y1	Y2	Y3	Y4	Y5
1-Apr-09	1.85	0.8	4.62		5.46
2-Apr-09	2.12	9.8	7.09		1.61
6-Apr-09	-3.14	2.5	11.63		10.48
8-Apr-09	1.76	4.39	-0.97		-4.23
9-Apr-09	-0.68	11.09	2.22		2.69
13-Apr-09	0.75	2.63	1.76		1.34
15-Apr-09	3.85	4.22	7.12		4.25
16-Apr-09	0.36	-5.08	-5.17		-4.33
17-Apr-09	-0.9	2.74	0.17		-1.18
20-Apr-09	1.25	-0.53	0.17		0.71
21-Apr-09	-0.82	3.74	2.78		0.24
22-Apr-09	5.26	-3.52	1.03		2.4
23-Apr-09	0.3	-0.15	6.44		0.6
24-Apr-09	4.88	0	14.55		3.36
27-Apr-09	-2.98	0.77	-4.29		-3.61
28-Apr-09	-2.32	-3.92	-7.17		-5.24
29-Apr-09	1.44	2.95	6.54		4.61
4-May-09	0.38	3.22	9.99		4.88
5-May-09	-0.21	-0.8	-2.37		1.53
6-May-09	-2.09	0.09	-1.31		-2.1
7-May-09	-2.44	0.09	1.03		2
8-May-09	-3.55	-2.33	-4.47		0.77

[robsonandrade0]

I made a program that I hope will help solve your problem.

data exemplo;
input Date $ 1-7 Y1 $ 8-13 Y2 $ 14-19 Y3 $ 20-25 Y4 $ 26-26 Y5 $ 27-33;
datalines;
1Apr09 1.85 0.8 4.62 5.46
2Apr09 2.12 9.8 7.09 1.61
6Apr09 -3.14 2.5 11.63 10.48
8Apr09 1.76 4.39 -0.97 -4.23
9Apr09 -0.68 11.09 2.22 2.69
13Apr09 0.75 2.63 1.76 1.34
15Apr09 3.85 4.22 7.12 4.25
16Apr09 0.36 -5.08 -5.17 -4.33
17Apr09 -0.9 2.74 0.17 -1.18
20Apr09 1.25 -0.53 0.17 0.71
21Apr09 -0.82 3.74 2.78 0.24
22Apr09 5.26 -3.52 1.03 2.4
23Apr09 0.3 -0.15 6.44 0.6
24Apr09 4.88 0 14.55 3.36
27Apr09 -2.98 0.77 -4.29 -3.61
28Apr09 -2.32 -3.92 -7.17 -5.24
29Apr09 1.44 2.95 6.54 4.61
4May09 0.38 3.22 9.99 4.88
5May09 -0.21 -0.8 -2.37 1.53
6May09 -2.09 0.09 -1.31 -2.1
7May09 -2.44 0.09 1.03 2
8May09 -3.55 -2.33 -4.47 0.77
;

data teste (drop= i Y1-Y5);
set work.exemplo;
array X{5} X1-X5;
array Y{5} Y1-Y5;
do i=1 to 5;
X{i}=Y{i}*1;
end;
RUN;

[robsonandrade0]

I made a program that I hope will help solve your problem.

data exemplo;
input Date $ 1-7 Y1 $ 8-13 Y2 $ 14-19 Y3 $ 20-25 Y4 $ 26-26 Y5 $ 27-33;
datalines;
1Apr09 1.85 0.8 4.62 5.46
2Apr09 2.12 9.8 7.09 1.61
6Apr09 -3.14 2.5 11.63 10.48
8Apr09 1.76 4.39 -0.97 -4.23
9Apr09 -0.68 11.09 2.22 2.69
13Apr09 0.75 2.63 1.76 1.34
15Apr09 3.85 4.22 7.12 4.25
16Apr09 0.36 -5.08 -5.17 -4.33
17Apr09 -0.9 2.74 0.17 -1.18
20Apr09 1.25 -0.53 0.17 0.71
21Apr09 -0.82 3.74 2.78 0.24
22Apr09 5.26 -3.52 1.03 2.4
23Apr09 0.3 -0.15 6.44 0.6
24Apr09 4.88 0 14.55 3.36
27Apr09 -2.98 0.77 -4.29 -3.61
28Apr09 -2.32 -3.92 -7.17 -5.24
29Apr09 1.44 2.95 6.54 4.61
4May09 0.38 3.22 9.99 4.88
5May09 -0.21 -0.8 -2.37 1.53
6May09 -2.09 0.09 -1.31 -2.1
7May09 -2.44 0.09 1.03 2
8May09 -3.55 -2.33 -4.47 0.77
;

data teste (drop= i Y1-Y5);
set work.exemplo;
array X{5} X1-X5;
array Y{5} Y1-Y5;
do i=1 to 5;
X{i}=Y{i}*1;
end;
RUN;