[R] how to calcaulate matrices for two subsets

[Adam D. I. Kramer]

Hi Bill,

Tell me more about the Obs object. The "subset" of an lm should be a vector
telling the lm which observations to use...if Obs[197,396] is a single
number, only one observation will be used, and chances are your model is not
what you intended.

Also, predict(result1,newdata=Obs[397,339]) needs that cell in Obs to have a
column called "SP500 index," and predict.lm (check the man page for it) will
only return one number, the prediction for the given new data.

If you want it to have 3 numbers, maybe Obs[397:399] is what you want? The
colon means "397 through 399, inclusive" while the comma means "row 397,
column 399".  If this is your mistake, you actually want "subset=197:396,
data=Obs" in your call to lm.

Hope this helped.

--Adam

On Thu, 11 Sep 2008, Xianchun Liao wrote:

> I am an R beginner and trying to run a market model using event study in
> R framework.
> 
> 
> 
> First, I run a market model, that is lm(stock security~SP500 index,
> subset=Obs[197, 396]) ->result1
> 
> Then I get predict results for a  new dataset using predict (result1,
> newdata=Obs[397,399]) ->pred1
> 
> Pred1 should have three numbers.
> 
> 
> 
> Now I need to calculate abnormal return by the formula stock security
> [397,399] -pred1
> 
> But it does not work after trying many times.
> 
> 
> 
> So, how to write a R code to implement the formula and get right
> results.
> 
> 
> 
> 
> 
> Thanks,
> 
> 
> 
> Bill
> 
>
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at 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.
>