search

refunders / refund

Regression with functional data
40 stars 23 forks source link

Problems to plot pffr objects #91

Open neotis70 opened 5 years ago

neotis70 commented 5 years ago

I tried to plot the result of a pffr I made. It was a function-on-scalar regression with the constant 1 and the sex as covariates. When I plotted it, the scale of the first coefficient was too high for the second one so I stored the results, and I plotted only the sex coefficient, using plot[[2]]$x as x, as.numeric(plot[[2]]$fit) +- 1.96 * plot[[2]]$se as y, and I got something not related to the data. The initial yind I use in the pffr are specifically created for my problem (50 points between 0 and 0.03, and 50 points between 0.03 and 1) and I'm wondering if the function plot.pffr takes it into account ?

fabian-s commented 5 years ago

Thanks for your report. Please provide a reproducible example of the problem so we can try to help you.