[R] help me to debug this part of code?

[Charles C. Berry]

On Fri, 18 Apr 2008, jinjin wrote:

>
> I am trying to solve the integration equation, for different values of K from
> 4 to 25, the integration is with respect to u,
>
> Here is the equation:    gamma(k/2) / ( sqrt(k-1)*gamma((k-1)/2) ) *
> integrate(f= (1+u^2/k-1)^(-k/2), lower=0, upper= sqrt(a^2*k/(k+1-a^2)) ) =
> the similar expression as te left hand except k becomes k+1
>
>
> my code is below, I don't know why R keep telling me the syntax wrong but I
> am not clear, thanks.

First, you should format your code so that it is more readable. Use spaces 
and indentation.

Second, it is recommended that use an editor that features syntax 
highlighting to prepare your code. It will show you which parentheses and 
braces match. And if it features automatic indentation, this will give you 
clues when the nesting of an expression or arglist has gone awry.

Third, read the error messages that R provides and try to understand why R 
is telling you "unexpected symbol", etc. Often, an error immediately 
preceeded the point at which R got confused.

If all this fails, break apart your code into single expressions (or 
pieces you ***think*** are valid expressions) and try to evaluate them. 
For example,

 	 m<-exp(lgamma(k/2)-lgamma((k-1)/2))*1/sqrt(k-1)

is correctly parsed. It only fails for me because I tried to evaluate it 
without providing a value for 'k'.

HTH,

Chuck

>
>
> f<-function(u,k){(1+u^2/(k-1))^(-k/2)}
> g<-function(u,k){(1+u^2/k)^(-(k+1)/2)}
>
> for(k in 4:25){
> out<-uniroot(function(a){
> m<-exp(lgamma(k/2)-lgamma((k-1)/2))*1/sqrt(k-1)
> n<-exp(lgamma((k+1)/2)-lgamma(k/2)*1/sqrt(k)
> m*integrate(f,lower=0,upper=sqrt(a^2*(k-1)/(k-a^2)),rel.tol=.Machine$double.eps^0.25)
> -n*integrate(g,lower=0,upper=sqrt(a^2*k/(k+1-a^2)),rel.tol=.Machine$double.eps^0.25)
> }
> lower=1.4, upper=2)
> print(unlist(out))}
>
>
> -- 
> View this message in context: http://www.nabble.com/help-me-to-debug-this-part-of-code--tp16763482p16763482.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

Charles C. Berry                            (858) 534-2098
                                             Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	            UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego 92093-0901