Dear R-helpers,
i have a problem with a glm-model. I am trying to fit models with the log as
link function instead of the logit. However, in some cases glm fails to
estimate those models and suggests to give start values. However, when I set
start = coef(logistic_model) within the function call, glm still says it cannot
find starting values? This seems to be more of a problem, when I include a
continous predictor in the model (age instead of group). find below a minimal
example.
Do I need to set other/better starting values? I would greatly appreciate any
hints!
Best, Felix
x = structure(list(Alter = c(28, 72, 48, 53, 49, 56, 47, 20, 72,
26, 28, 28, 25, 63, 42, 23, 68, 63, 44, 23, 23,
47, 30, 22, 21,
30, 26, 47, 40, 43, 23, 78, 29, 20, 49, 70, 24,
49, 43, 49, 68,
50, 42, 27, 70, 68, 46, 42, 40, 44, 48, 24, 23,
24, 56, 60, 66,
40, 71, 45, 37, 71, 41, 53, 48, 34, 52, 26, 76,
46, 65, 69, 75,
59, 30, 54, 69, 46, 50, 62, 38, 34, 30, 29, 73,
20, 57, 64, 40,
28, 21, 36, 65, 22, 69, 24, 38, 61, 70, 47, 61,
20, 58, 29, 35,
23, 29, 22, 21, 56, 37, 79, 27, 25, 75, 64, 22,
48, 36, 24, 44,
38, 23, 54, 76, 43, 30, 47, 48, 23, 68, 28, 44,
54, 43, 35, 47,
49, 44, 53, 26, 24, 56, 34, 39, 67, 74, 49, 55,
39, 58, 69, 46,
56, 69, 69, 26, 58, 41, 46, 40, 49, 24, 29, 24,
71, 41, 61, 27,
25, 38, 56, 26, 53, 39, 77, 40, 53, 61, 61, 54,
62, 28, 71, 42,
67, 44, 20, 40, 27, 27, 22, 71, 24, 31, 63, 24,
22, 30, 42, 43,
23, 46, 49, 21, 25, 30, 64, 29, 52, 29, 50, 57,
50, 53, 50, 34,
58, 42, 35, 50, 35, 35, 63, 42, 37, 64, 34, 56,
70, 48, 23, 43,
26, 52, 24, 31, 27, 34, 23, 44, 51, 41, 69, 47,
37, 68, 42, 28,
25), Arthrose = structure(c(1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 2L,
1L, 1L, 1L, 1L, 2L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 1L,
1L, 1L, 1L, 1L, 1L,
2L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 2L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 2L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 2L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 2L, 1L, 2L, 2L, 1L, 1L, 1L,
2L, 2L, 1L, 1L, 2L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
2L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 1L,
1L), .Label = c("nicht
erkrankt", "erkrankt"), class = "factor"),
Gruppe = c(2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1,
1, 1, 2, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2,
2, 2, 2, 2, 2, 2)), .Names = c("Alter",
"Arthrose", "Gruppe"
), row.names = c(6L, 8L, 49L, 53L, 54L, 84L, 87L,
88L, 110L,
139L, 145L, 156L, 167L, 176L,
177L, 178L, 189L, 193L, 216L, 237L,
245L, 269L, 272L, 280L, 303L,
314L, 315L, 326L, 338L, 344L, 345L,
352L, 365L, 366L, 377L, 393L,
401L, 404L, 409L, 439L, 469L, 505L,
510L, 544L, 552L, 559L, 561L,
586L, 597L, 598L, 601L, 607L, 611L,
630L, 650L, 663L, 672L, 673L,
689L, 690L, 719L, 747L, 794L, 809L,
818L, 819L, 840L, 869L, 878L,
886L, 905L, 913L, 915L, 924L, 937L,
955L, 963L, 970L, 978L, 985L,
997L, 1005L, 1021L, 1022L, 1033L,
1040L, 1041L, 1043L, 1066L,
1068L, 1084L, 1099L, 1112L, 1113L,
1125L, 1134L, 1154L, 1155L,
1166L, 1171L, 1195L, 1208L, 1216L,
1217L, 1229L, 1230L, 1236L,
1242L, 1252L, 1288L, 1308L, 1360L,
1365L, 1371L, 1383L, 1384L,
1402L, 1406L, 1412L, 1413L, 1438L,
1448L, 1451L, 1455L, 1459L,
1478L, 1483L, 1492L, 1508L, 1511L,
1519L, 1531L, 1554L, 1569L,
1573L, 1590L, 1615L, 1629L, 1649L,
1651L, 1654L, 1660L, 1661L,
1674L, 1684L, 1687L, 1690L, 1696L,
1724L, 1730L, 1767L, 1775L,
1779L, 1780L, 1800L, 1801L, 1829L,
1837L, 1848L, 1884L, 1909L,
1916L, 1933L, 1934L, 1952L, 1970L,
1991L, 2021L, 2024L, 2029L,
2040L, 2060L, 2095L, 2112L, 2115L,
2122L, 2131L, 2145L, 2150L,
2173L, 2188L, 2189L, 2193L, 2197L,
2240L, 2251L, 2252L, 2264L,
2266L, 2277L, 2313L, 2315L, 2318L,
2324L, 2331L, 2336L, 2344L,
2345L, 2357L, 2366L, 2384L, 2392L,
2413L, 2422L, 2453L, 2474L,
2477L, 2480L, 2484L, 2499L, 2502L,
2518L, 2548L, 2551L, 2565L,
2575L, 2584L, 2607L, 2608L, 2617L,
2620L, 2644L, 2653L, 2654L,
2655L, 2667L, 2669L, 2672L, 2686L,
2697L, 2733L, 2739L, 2742L,
2750L, 2764L, 2774L, 2783L, 2787L,
2807L, 2817L, 2841L, 2847L,
2850L, 2852L, 2860L, 2863L, 2889L,
2908L, 2917L, 2924L), class =
"data.frame")
Group_logit_model = glm(data = x, Arthrose ~ Gruppe, family=binomial(link =
logit))
Group_log_model = glm(data = x, Arthrose ~ Gruppe, family=binomial(link = log))
Age_logit_model = glm(data = x, Arthrose ~ Alter, family=binomial(link = logit))
Age_log_model = glm(data = x, Arthrose ~ Alter, family=binomial(link = log))
Age_log_model_start = glm(data = x, start = coef(Age_logit_model), Arthrose ~
Alter, family=binomial(link = log))
Dr. rer. nat. Felix Fischer
Diplom-Psychologe
Institut für Sozialmedizin, Epidemiologie und Gesundheitsökonomie
Charité - Universitätsmedizin Berlin
Luisenstrasse 57
10117 Berlin
Tel: 030 450 529 104
Fax: 030 450 529 902
http://epidemiologie.charite.de
______________________________________________
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.
