[R-sig-eco] comparing results of var part

Hi all I am working in some variation partiting analysis using two different sets of predictive matrices but with the same set of explanatory variables. As expected I end up getting the results for % variation explained by [E], [S], [E+S] and unexplained variation. The results I get for both

[R-sig-eco] Help for the function "Simplex" in R

Hello I am a new user of R and try to use Simplex to solve the linear programming problem. My problem has 47 variable. enj is the vector of object function which is 1*47 dimension M1 and M2 are 2 1000*48 dimension matrixes. The first column in M1 indicated b1 and 2-48 for the A1 as following

[R-sig-eco] comparing results of varpart

Hi all I am working in some variation partiting analysis using two different sets of predictive matrices but with the same set of explanatory variables. As expected I end up getting the results for % variation explained by [E], [S], [E+S] and unexplained variation. The results I get for both

Re: [R-sig-eco] Analyzing frequencies in R

Dear Manuel, You can use the Wilson method (with Yates' continuity correction) to calculate CIs for proportion data. It's formally described and advocated in the following articles: Newcombe R.G. (1998) Two-Sided Confidence Intervals for the Single Proportion: Comparison of Seven Methods.

[R-sig-eco] Analyzing frequencies in R

Dear list members, If I have 3 frequencies (3 mutually exclusive groups): white: 19 black: 43 red: 24 How can I obtain confidence intervals for the proportions, instead of a P value from a chisquare test in R? Or better, how can I assess "effect size" instead of finishing the analysis on a P

Re: [R-sig-eco] plotting numeric v. factor variables

Bessie, Another, non-trellis/lattice way to do this is simply with boxplot(). boxplot(response~factor) will plot a boxplot for each of your factor levels. For greater control (i.e. removing whiskers, etc.) of boxplot see the help on boxplot(), bxp(), and boxplot.stats(). Hope this is along the