Using stored scalar

[Zuhumnan Dapel]

Dear All,
I stored a scalar named, inv_a How do I use it if I want to run an estimation say

nl( y=({alpha=inv_a}^5 )*l5.y

?

Thanks

[Rich Goldstein]

have a look at the manual: [P] scalar; the easy way to get there is "h scalar" and then click on the link at the top of the help file

[Zuhumnan Dapel]

Thanks Rich. However I could't get related help

[Carlo Lazzaro]

Zuhumnan:
I do hope that what follows may be helpful:

use auto.dta, clear
scalar a=2
g y=runiform()
g x=runiform()
nl (y = {a} + { weight }*x^{ turn =1})

[Zuhumnan Dapel]

Thanks Carlo. However, the idea is that I must associate the parameter being estimated with an initial value. For instance, am estimating alpha and its initial value is stored as a scalar with the name inv_a. Therefore the user function has to be

  
 nl( y=({alpha=inv_a}^5 )*l5.y

. It works if I use

  
 nl( y=({alpha=0.7}^5 )*l5.y

. I prefer the first option because I have many of such equations to estimate.

[Nick Cox]

nl( y=({alpha=`=inv_a'}^5 )*l5.y

[Zuhumnan Dapel]

Goodness me! Dr Cox you are such an amazing person on this forum. My longstanding problem is now solved. Thanks a million.