Dear R
expert
I am a
student and I am currently conducting a research project on the Modeling Loss
Index Triggers to price Cat Bonds: Application of the risk of hurricanes in
USA.
I need to
solve with R (especially with EM algorithm) this specific problem below. CRAN
Package archive doesn't seem to have it also the statistical modeling journal
didn't contain a paper that implements this:
I have two
data samples. The first sample contains 64 depths for historical hurricanes
happened in Florida between 1950 - 2008. The second sample contains only 48
losses associated to these hurricanes. This second data is missing 16 values of
losses.
Since 48
out 64 observations have information about hurricanes losses, it is necessary
to treat the missing data of losses for a further analysis. So, I chose to make
an EM-algorithm:
1- Experts in climatology describe that
hurricanes losses are directly proportional to the depths of the hurricanes.
Besides, statistically we observe a relationship between the two vectors
(depths of hurricanes and losses). So the approach is to estimate the missing
losses by means of the linear regression model x= á+ â yi + åi, where x (n x 1)
is a vector of observations
on the response variable and y (n x p)
is the p explanatory variables.
2- An approach to deal with the missing data
is the expectation -Maximum algorithm (EM)
The
expectation step:
This
algorithm consists of omitting the cases with missing data and running a
regression on what remains. The
regression coefficient will be used to estimate the missing data.
The
maximization step:
After this
estimation step, a new regression will be done over the complete data
(including estimated values). With the new regression coefficients, the missing
data is re-estimated. This process will continue until the estimates are
adjusted to give model sampling error, ie it will not be a longer noticeable
change.
Does anyone
have any ideas how to make this in R algorithm?
Can you
guide me to something already done? Or please help me to find the true code.
Thank you
in advance to consider my request
Best regards
Souad
_________________________________________________________________
[[elided Hotmail spam]]
[[alternative HTML version deleted]]
______________________________________________
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.
