I need help in interpreting some output of polymars. The returned model by my command is:
m$model pred1 knot1 pred2 knot2 coefs SE 1 0 NA 0 NA 1.32222163 0.0007806758 2 1 NA 0 NA -0.10904285 0.0006735827 3 1 1.193575 0 NA 1.25396217 0.0205092438 4 1 1.205400 0 NA -0.18020096 0.0261931173 5 1 1.230725 0 NA 0.04709508 0.0118559207 6 1 1.285400 0 NA -0.02593593 0.0097733958 7 1 1.317275 0 NA 0.03516571 0.0177043207 I'm trying to confirm the generated output of mars by manually entering the basis functions into my dataset, What I don't understand is that in the knot1 column there are to rows containing NA. In the polymars literature I was reading this means that these basis functions are linear. BF1 = MAX(0, A2 - 1.193575) BF2 =MAX(0, 1.193575 - A2) BF3 = MAX(0, A2 - 1.2054) BF5 = MAX(0, A2 - 1.230725) BF7 = MAX(0, A2 - 1.2854) BF11 = MAX(0, A2 - 1.317275) RESPONSE = 1.32222163 - 0.10904285 * C2 + 1.25396217 * D2 - 0.18020096 * E2 + 0.04709508 * F2 - 0.02593593 * G2 + 0.03516571 * H2 On the attached, the CLOSE column is my source data, the ESTIMATE is the column generated by polymars, the PREDICTION column is my formula. Can I have some help with this please.
______________________________________________ R-help@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.
