Dear colleagues, I have the following dataset. It is modelled on the data
included in Box-Seteffenheiser and Jones "Event History Modelling"
Using the following code, I try to find the baseline hazard function
haz_1<-muhaz(bpa$time, bpa$censored, subset=(bpa$year=="2010" |
bpa$ban=="1"), min.time=1, max.time=3)
I think I'm doing everything right, but what I don't understand is how
to derive a duration dependency coefficient rom the values contained in the
muhaz object as per Box-Steffenheiser and Jones' recommendations in Ch. 5 of
Event History Modelling.
I get the following summary(haz_1)
Number of Observations .......... 50
Censored Observations ........... 43
Method used ..................... Local
Boundary Correction Type ........ Left and Right
Kernel type ..................... Epanechnikov
Minimum Time .................... 1
Maximum Time .................... 3
Number of minimization points ... 51
Number of estimation points ..... 101
Pilot Bandwidth ................. 0.25
Smoothing Bandwidth ............. 1.27
Minimum IMSE .................... 6716.9
Can anyone provide any advice?
Yours, Simon Kiss
'data.frame': 147 obs. of 7 variables:
$ state : Factor w/ 50 levels "Alabama","Alaska",..: 1 1
1 2 2 2 3 3 3 4 ...
$ partisan: Factor w/ 3 levels "democrat","mixed",..: 1 1 1
2 2 2 3 3 3 1 ...
$ ban : num 0 0 0 0 0 0 0 0 0 0 ...
$ year : num 2008 2009 2010 2008 2009 ...
$ news : num 1.67 1.67 0 2 0 ...
$ time : num 1 2 3 1 2 3 1 2 3 1 ...
$ censored: num 0 0 0 0 0 0 0 0 0 0 ...
*********************************
Simon J. Kiss, PhD
Assistant Professor, Wilfrid Laurier University
73 George Street
Brantford, Ontario, Canada
N3T 2C9
Cell: +1 519 761 7606
