s-news
[Top] [All Lists]

Problem with nlme and Sslogis and SSfpl

To: "'S-News (s-news@lists.biostat.wustl.edu)'" <s-news@lists.biostat.wustl.edu>
Subject: Problem with nlme and Sslogis and SSfpl
From: Vadim Kutsyy <vkutsyy@cytokinetics.com>
Date: Fri, 14 Feb 2003 10:14:11 -0800
I am having problem running nlme with Logistic model  and Four-parameter
Logistic Model.  Splus gives me a fitting error.  Any recommendations?

Here is Splus output of a simple model:

-----------------------------------
> fit<-nlme(y~SSfpl(logx,A, B, xmid, scal),
+           data=z,
+           fixed=A+ B+ xmid+ scal~1
+       )
Problem in nls(y ~ cbind(1, 1/(1 + exp((xmid - x)/exp(lscal)))), data =
data.frame(xy..: step factor reduced below minimum 
Use traceback() to see the call stack

> fit<-nlme(y~SSlogis(logx,Asym, xmid, scal),
+           data=z,
+           fixed=Asym+ xmid+ scal~1
+       )
Problem in nls(y ~ 1/(1 + exp((xmid - x)/scal)), data = xy, start =
list(xmid = aux[1..: step factor reduced below minimum 
Use traceback() to see the call stack

--------------------------------

And here is dump of z:

------------------------------------------
"z" <-
  structure(.Data = list(c(-6.1508755397365853, -5.9004574713046667, 
              -5.6500390210702003, -5.3996196291291731, -5.1491994308754458,
-4.8987796212089441,
              -4.3979400086720375, -6.1508755397365853, -5.9004574713046667,

              -5.6500390210702003, -5.3996196291291731, -5.1491994308754458,
              -4.8987796212089441, -4.6483598198914011, -4.3979400086720375,

              -6.1508755397365853, -5.9004574713046667, -5.6500390210702003,
              -5.3996196291291731, -5.1491994308754458, -4.8987796212089441,

              -4.6483598198914011, -4.3979400086720375)
              , c("604" = 0.047212741751990896
                  , "588" = 0.05128205128205128
                  , "603" = 0.029541689674213143
                  , "587" = 0.029633933759442184
                  , "602" = 0.048735511064278189
                  , "586" = 0.057489123679303915
                  , "585" = 0.1094941094941095
                  , "143" = 0.034290738569753815
                  , "123" = 0.031791907514450865
                  , "142" = 0.055234239444765756
                  , "122" = 0.040374488004681103
                  , "141" = 0.0344336937962436
                  , "121" = 0.023044269254093388
                  , "140" = 0.033013205282112844
                  , "120" = 0.19625137816979052
                  , "376" = 0.021447721179624665
                  , "356" = 0.026430348258706465
                  , "375" = 0.02303894679100384
                  , "355" = 0.024709302325581398
                  , "374" = 0.03775811209439528
                  , "354" = 0.028307692307692308
                  , "373" = 0.03327283726557774
                  , "353" = 0.13589128697042369
                  )
              )
            , names = c("logx", "y")
            , class = "data.frame"
            , row.names = c("141", "125", "140", "124", "139", "123", "122",
                "374", "354", "373", "353", "372", "352", "371", "351",
"607", "587", "606", "586", "605",
                "585", "604", "584")
            )


===================================================================
Vadim Kutsyy, PhD                             http://www.kutsyy.com
vkutsyy@cytokinetics.com                           vadim@kutsyy.com
Statistician                                      tel. 650-624-3218
Cytokinetics, Inc                                 fax. 650-624-3010
280 East Grand Avenue
South San Francisco, CA 94080           http://www.cytokinetics.com
===================================================================  

<Prev in Thread] Current Thread [Next in Thread>
  • Problem with nlme and Sslogis and SSfpl, Vadim Kutsyy <=