Display Format

[Sattar Khan]

Greetings Everyone!

Hopefully, you all be doing well, I have a question regarding the display format for a table i.e

pwcorr MCDA REALEM DIS_ERM RCFO PROD_RM BIND BEXP BF FD

             |     MCDA   REALEM  DIS_ERM     RCFO  PROD_RM     BIND     BEXP
-------------+---------------------------------------------------------------
        MCDA |   1.0000 
      REALEM |  -0.0598   1.0000 
     DIS_ERM |  -0.0059   0.2531   1.0000 
        RCFO |  -0.2504   0.7301  -0.0068   1.0000 
     PROD_RM |   0.0398   0.8289   0.0069   0.5848   1.0000 
        BIND |   0.0190  -0.0958   0.0181  -0.0683  -0.1203   1.0000 
        BEXP |  -0.0310  -0.1156  -0.0043  -0.0692  -0.1164   0.1972   1.0000 
          BF |  -0.0020   0.0694   0.0205   0.0687   0.0301  -0.0472  -0.1783 
          FD |   0.0045   0.1195  -0.0089   0.0728   0.1436  -0.1967  -0.0021 

             |       BF       FD
-------------+------------------
          BF |   1.0000 
          FD |   0.2440   1.0000

As shown in the table that there is 4 digit after the decimal, I want to reduce it to two-digit, in regression we do that by b(2) t(2), How we will do it in this situation?

Thanks in Advance

[Sergiy Radyakin]

Try:

sysuse auto
pwcorr price length weight // can't specify format here
matrix list r(C), format(%6.2f) // can specify format here

[Sattar Khan]

Try:

sysuse auto
pwcorr price length weight // can't specify format here
matrix list r(C), format(%6.2f) // can specify format here

Kindy Sergiy elaborate it , because it is not working

[Maarten Buis]

It is not working, is not enough information for us to identify what the problem is. Besides the code given by Sergiy works just fine:

. sysuse auto
(1978 automobile data)

. pwcorr price length weight // can't specify format here

             |    price   length   weight
-------------+---------------------------
       price |   1.0000
      length |   0.4318   1.0000
      weight |   0.5386   0.9460   1.0000

. matrix list r(C), format(%6.2f) // can specify format here

symmetric r(C)[3,3]
         price  length  weight
 price    1.00
length    0.43    1.00
weight    0.54    0.95    1.00

What did you do exactly (emphasis on exactly), and why do you think that it is not working?

[Sattar Khan]

It is not working, is not enough information for us to identify what the problem is. Besides the code given by Sergiy works just fine:

. sysuse auto
(1978 automobile data)

. pwcorr price length weight // can't specify format here

| price length weight
-------------+---------------------------
price | 1.0000
length | 0.4318 1.0000
weight | 0.5386 0.9460 1.0000

. matrix list r(C), format(%6.2f) // can specify format here

symmetric r(C)[3,3]
price length weight
price 1.00
length 0.43 1.00
weight 0.54 0.95 1.00

What did you do exactly (emphasis on exactly), and why do you think that it is not working?

Thanks for your comment Maarten, what should I write "can specify format here" after running the code the result was as follows

pwcorr MCDA REALEM DIS_ERM RCFO PROD_RM BIND BEXP BF FD

             |     MCDA   REALEM  DIS_ERM     RCFO  PROD_RM     BIND     BEXP
-------------+---------------------------------------------------------------
        MCDA |   1.0000 
      REALEM |  -0.0598   1.0000 
     DIS_ERM |  -0.0059   0.2531   1.0000 
        RCFO |  -0.2504   0.7301  -0.0068   1.0000 
     PROD_RM |   0.0398   0.8289   0.0069   0.5848   1.0000 
        BIND |   0.0190  -0.0958   0.0181  -0.0683  -0.1203   1.0000 
        BEXP |  -0.0310  -0.1156  -0.0043  -0.0692  -0.1164   0.1972   1.0000 
          BF |  -0.0020   0.0694   0.0205   0.0687   0.0301  -0.0472  -0.1783 
          FD |   0.0045   0.1195  -0.0089   0.0728   0.1436  -0.1967  -0.0021 

             |       BF       FD
-------------+------------------
          BF |   1.0000 
          FD |   0.2440   1.0000 

. matrix list r(C), format(%6.2f)

symmetric r(C)[2,2]
      FD    FD
FD  1.00
FD  1.00  1.00

[Maarten Buis]

OK, so it looks like you have Stata 13 or less. This is a good illustration why it is so important that you tell us from the very beginning when you have an older version of Stata.

If you use corr instead of pwcorr, then it should work. corr is probably better than pwcorr anyhow for most applications: If you are displaying a correlation table in your paper, then you want the sample for that table to correspond to the sample used in the later analysis.

[Sattar Khan]

OK, so it looks like you have Stata 13 or less. This is a good illustration why it is so important that you tell us from the very beginning when you have an older version of Stata.

If you use corr instead of pwcorr, then it should work. corr is probably better than pwcorr anyhow for most applications: If you are displaying a correlation table in your paper, then you want the sample for that table to correspond to the sample used in the later analysis.

Thank You Mr. Maarten, I got it, it's work on corr.

[Nick Cox]

Although I agree with Maarten Buis, which is why I rarely use pwcorr rather than correlate, I note that a quick

1. copy pwcorr.ado to your space

2. edit 7.4f to 7.2f

3. change the program name and save it to a new file with the same name

seems to work. See also corrci from the Stata Journal.

[Sattar Khan]

Although I agree with Maarten Buis, which is why I rarely use pwcorr rather than correlate, I note that a quick

1. copy pwcorr.ado to your space

2. edit 7.4f to 7.2f

3. change the program name and save it to a new file with the same name

seems to work. See also corrci from the Stata Journal.

Thank you for your comment Mr. Nick