R help - Nov 2013 - How can I get seasonal variation table from generalized additive mixed models in R?

Hello everyone,

I used a function called gamm in mgcv in R program to investigate seasonal
variation of wind speed based on 13 years of measurement dataset. I would
like to use this variation in my formula for my study but as I am new in R,
it's extremely hard to find values on y-axis.. too many sub-directories of
gamm values make me hard to find what directory was actually used for
y-axis. 
My data set is like this below:

time /day.of.year/ month /day.of.month  /year   / WindSpeed
    1       1                7            24            2000       23.429
    2       2                7            25            2000       29.170
    3       3                7            26            2000       16.813
    4       4                7            27            2000       15.271
    5       5                7            28            2000       10.125
    6       6                7            29            2000       13.938
    7       7                7            30            2000       15.854
    8       8                7            31            2000       10.438
    9       9                8            1              2000       7.125
    .         .                .             .                .            .
    .         .                .             .                .            .

As time stamps are separated, I combined all of them first and then ran
additive modelling to look at seasonal and yearly trends of 13 years of wind
data.

My R codes are described below:
ballarat  <- read.csv("ballarat seasonal daily.csv", header=TRUE,
sep=",")
ballarat1 <- within(ballarat, Date <- as.Date(paste(year, month,
day.of.month, sep = "-")))

plot(WindSpeed ~ Date, data = ballarat1, type = "l")

mod       <- gamm(WindSpeed ~ s(day.of.year, bs = "cc") + s(time,
bs "cr"),
             data = ballarat1, method = "REML",
             correlation = corAR1(form = ~ 1 | year),
             knots = list(day.of.year = c(0, 366)))

summary(mod$gam)

plot(mod$gam, pages = 1)


http://i.stack.imgur.com/wU0AU.jpg
>From the link page of trend graphs, You can see that average wind speed has
about +- 2 km/hr variation. I would like to produce tables (two columns
(first column: x axis values, second column: y axis (variations)) each day
of year) of those two graphs. I assume I need to use gam.predict to get
tables of the prediction? As I am new in R, I still can't understand how I
can make it work even after reading description of ?gam.predict function.
Could you please show me how to do it step by step?

Regards,

Kangmin.





--
View this message in context:
http://r.789695.n4.nabble.com/How-can-I-get-seasonal-variation-table-from-generalized-additive-mixed-models-in-R-tp4680710.html
Sent from the R help mailing list archive at Nabble.com.