Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Re: Using NEARMRG

From   "MA V" <[email protected]>
To   [email protected]
Subject   Re: st: Re: Using NEARMRG
Date   Wed, 26 Apr 2006 23:02:27 +0000

Yes, you're right... I had 1 (!!) missing value...
Thanks again.

From: "Michael Blasnik" <[email protected]>
Reply-To: [email protected]
To: <[email protected]>
Subject: Re: st: Re: Using NEARMRG
Date: Wed, 26 Apr 2006 18:55:26 -0400

-nearmrg- calls the Stata command -isid- to check that the variable list uniquely identifies the observations. Have you looked at the data to see if there are duplicates? Perhaps it's a problem with missing values since isid will not allow missing values wihtout specifying the missok option. If that is the problem, I suggest you drop the cases with missing values on either age or region.

----- Original Message ----- From: "MA V" <[email protected]>
To: <[email protected]>
Sent: Wednesday, April 26, 2006 6:44 PM
Subject: RE: st: Re: Using NEARMRG

I know you're the author and I thank you for these and many other programs that have helped me a lot.
One of the things I tried was dropping all the duplicates for both region and age in the using file (before doing the merge):
duplicates drop age, force
duplicates drop region, force

This should have made the nearvar unique in the using dataset within levels of the varlist, right?

But I keep getting the same error message.

Another thing that I forgot to mention is that, in my case age is a string variable (and yes, I used the option "upper" to prevent getting error messages related with the fact that the nearvar is a string variable).

If you have any suggestions I would appreciate your help.

*   For searches and help try:
Express yourself instantly with MSN Messenger! Download today - it's FREE!

* For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index