Re: [R] Correlation Matrix with a Covariate

2014-09-02 Thread Patzelt, Edward
Even reducing the command to less variables than observations I still get: set.cor(y = (66:76), x = c(1:6), z = 65, data = dat5) Error in solve.default(x.matrix, xy.matrix) : Lapack routine dgesv: system is exactly singular: U[2,2] = 0 On Mon, Sep 1, 2014 at 12:25 PM, David L Carlson wrote:

Re: [R] Correlation Matrix with a Covariate

2014-09-02 Thread David L Carlson
[R] Correlation Matrix with a Covariate Even reducing the command to less variables than observations I still get: set.cor(y = (66:76), x = c(1:6), z = 65, data = dat5) Error in solve.default(x.matrix, xy.matrix) :    Lapack routine dgesv: system is exactly singular: U[2,2] = 0 On Mon, Sep 1, 2014 at

Re: [R] Correlation Matrix with a Covariate

2014-09-01 Thread David L Carlson
Thanks for including your data with dput(). I'm not familiar with set correlation, but altogether you are working with 76 variables (columns) and only 46 observations. Since the error message says "the system is exactly singlular," it is likely that you have too many variables for the number of