Please bear with me, I am very new to R. My question is regarding the use of the improveProb function in the Hmisc package. I have two logistic models, the only difference being that the second model contains my novel marker of interest. I am trying to calculate NRI and IDI to compare models. I have the PredRisks for both models - PredRisk1 and PredRisk2, and my outcome is disease 0/1. How do I define this in R in order to run improveProb(x1, x2, y)? Many thanks in advance -- View this message in context: http://r.789695.n4.nabble.com/Help-with-improveProb-function-in-Hmisc-in-R-tp4713004.html Sent from the R help mailing list archive at Nabble.com.
On Oct 1, 2015, at 2:26 AM, kirsada wrote:> Please bear with me, I am very new to R. > > My question is regarding the use of the improveProb function in the Hmisc > package. I have two logistic models, the only difference being that the > second model contains my novel marker of interest. I am trying to calculate > NRI and IDI to compare models. > > I have the PredRisks for both models - PredRisk1 and PredRisk2, and my > outcome is disease 0/1. How do I define this in R in order to run > > improveProb(x1, x2, y)?If you are saying that you have two numerical values each of length one, than I do not think you can pass those to a function that expects raw data. If your "PredRisk" variables are vectors of the same length as y and having a range of [0,1] as befits a probability, then what problems are you experiencing? -- David Winsemius Alameda, CA, USA
Hi David, Thanks for your response. I have discovered that my first and second models differed in length, and have since fixed them. Kirsten -- View this message in context: http://r.789695.n4.nabble.com/Help-with-improveProb-function-in-Hmisc-in-R-tp4713004p4713053.html Sent from the R help mailing list archive at Nabble.com.