Re: st: superposed rolled windows

["Michael S. Hanson" <mshanson@mac.com>]

On Jun 27, 2006, at 12:05 PM, irodriguez wrote:

> Michael, in your command:
>    tsset Firms Years     // Declare data as a panel
>    forval i = 1/3 {
>          local j = `i'-1
>         gen Profit`i' = F`j'.Profit
>     }
>    drop if missing(Profit3)
>
> what is the significate of 1/3, because in the reality i need to make
> superposed rolled windows for 5 years and I want to know if I have  
> to change the command and consider 1/5
	Yes, the "3" in "1/3" represents the number of years you specified  
in your earlier message.  You could try the following:


tsset Firms Years     // Declare data as a panel
local m 5             // Maximum number of years for "rolled windows"
forval i = 1/`m' {
    local j = `i'-1
    gen Profit`i' = F`j'.Profit
}
drop if missing(Profit`m')


	Notice that in addition to the -local- command, the -forval- and - 
drop- commands have changed vis-a-vis the quoted example.  If you  
decide to go back to a 3-year "superposed rolled windows" (in  
increase it to, say, 6), you can just change the "5" in the second  
line to the desired number of years, and the rest of your code will  
run as intended.  (At least as far as I understand what you intend.)


On Jun 27, 2006, at 12:14 PM, irodriguez wrote:


> When I try
> . tsset empresa ao
> stata say
> varlist:  empresa:  string variable not allowed
> I have tried with destring command but I don`t know why the data  
> are still
> string
	Are you sure the data are (now) numeric?  -desc empresa- will tell  
you if they are int(egers) or str(ings).  Without knowing more about  
your data, it is difficult to know if what you really want is - 
encode- rather than -destring-.

	More generally, I encourage you to follow Nick Cox's dictum[*] of  
explaining exactly what you want to do, and listing the exact  
commands you have typed in your attempt to do so.  An actual (but  
brief) extract of your data can be helpful to those providing  
assistance as well.  HTH.

                                        -- Mike


[*] My attribution of this premise to Nick is (merely) empirical, as  
I have often observed him making similar comments in similar  
circumstances.  See the Statalist FAQ at <http://www.stata.com/ 
support/faqs/res/statalist.html> for a more expansive discussion of  
this issue.
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/