I am a new user of R. I am trying to fit gam model with our air pollution data. I used Foreign package to call data from SPSS and used MGCV package to fit gam. The following are the steps I used:> dust<- read.spss("a:dust9600jan.sav") > c<-gam(MRESPALL~s(DUSTM)+s(TEMP)+s(RH),family=poisson,data=dust) > summary(c)Family: poisson Link function: log Formula: MRESPALL ~ s(DUSTM) + s(TEMP) + s(RH) Parametric coefficients: Estimate std. err. t ratio Pr(>|t|) constant -0.61508 0.03331 -18.47 < 2.22e-16 Approximate significance of smooth terms: edf chi.sq p-value s(DUSTM) 6.605 16.738 0.01515 s(TEMP) 2.47 5.253 0.10705 s(RH) 1 1.5727 0.20981 R-sq.(adj) = 0.0202 Deviance explained = 2.35% UBRE score = 0.0031294 Scale est. = 1 n = 1692 I would like to know the relative risk. I don't know the command. Please, can anybody help me to get the relative risk. I am grateful for the help. Joseph E. Gomez [[alternative HTML version deleted]]
I am a new user of R. I am trying to fit gam model with our air pollution data. I used Foreign package to call data from SPSS and used MGCV package to fit gam. The following are the steps I used:> dust<- read.spss("a:dust9600jan.sav") > c<-gam(MRESPALL~s(DUSTM)+s(TEMP)+s(RH),family=poisson,data=dust) > summary(c)Family: poisson Link function: log Formula: MRESPALL ~ s(DUSTM) + s(TEMP) + s(RH) Parametric coefficients: Estimate std. err. t ratio Pr(>|t|) constant -0.61508 0.03331 -18.47 < 2.22e-16 Approximate significance of smooth terms: edf chi.sq p-value s(DUSTM) 6.605 16.738 0.01515 s(TEMP) 2.47 5.253 0.10705 s(RH) 1 1.5727 0.20981 R-sq.(adj) = 0.0202 Deviance explained = 2.35% UBRE score = 0.0031294 Scale est. = 1 n = 1692 I would like to know the relative risk. I don't know the command. Please, can anybody help me to get the relative risk. I will be grateful for the help. Joseph E. Gomez [[alternative HTML version deleted]]
Hi,
I wonder if it is possible to read data on another computer (unix like system)
using ssh (I mean any of ssh, scp, sftp). I saw that R has no problem reading
files across the net (for example read.table(url("http://www...")) ),
but what
if the file is on a local disk of another machine? For security reasons we are
not allowed to have the normal ftp (and telnet) running.
I played a bit with url(file://), but I don't know how (if it's
possible) to
specify both the address of the computer and the file name. I tried like this
url("file://host_name/path/filename") and
url("file://host_name:/path/filename")
but it didn't work.
I want to do this because my data files are big (500Mb) and I don'r want to
have
copies on every computer I use.
Thank you!
Giampiero