R help - Apr 2013 - Creating quintiles on monthly basis

Hi,

I am trying in R to indicate in which quintile a value of a variable is for
every month of my data frame in this case based on volatility. For each
month I want to know for each stock if it is in the most volatile quintile
of if it is in one of the others.

So far I have come up with the following function (see below).
Unfortunately, the function only works in some cases and often gives the
following error:

Error in cut.default(df$VOLATILITY, unique(breaks), label = FALSE, na.rm
=TRUE): 

  invalid number of intervals

Could you give me some advice on how to improve this code so that it works
properly.

It's relatively urgent. Many thanks!

quintilesVolByMonth <- function(x){

  months<-as.vector(unique(x$DATE))  

  dfx<-data.frame()

  for(n in seq(1,length(months))){

    num<-5

    print(paste("Appending month",months[n],sep=""))

    df<-subset(x,DATE==months[n])

    breaks<-quantile(df$VOLATILITY,probs=seq(0,1, 1/num),na.rm=TRUE)

    df$volquintile <- cut(df$VOLATILITY,unique(breaks), 

                       label=FALSE, na.rm=TRUE)

    dfx<-rbind(dfx,df)

  }

  return(dfx)

}

Frame.Quintile <- quintilesVolByMonth(x) 

 

 

Example of data: (I have 30 years of monthly data for over 1000 stocks)
Variable 'VolQuintile' is what I am trying to get.

 

DATE         STOCK     VOLATILITY     VolQuintile

01/01/1990         STOCK 1               0.3
1

01/01/1991         STOCK 2               0.15
4

01/05/1992         STOCK 3               0.1
3

.


	[[alternative HTML version deleted]]

Also posted at StackOverflow but now has an example dataset there.

Cedric: You are requested not to cross-post to Rhelp. There are many other
requests in the Posting Guide as well including a request that you not post in
HTML.

-- 
David.

On Apr 5, 2013, at 5:20 PM, Cedric Suykens wrote:
> Hi,
> 
> I am trying in R to indicate in which quintile a value of a variable is for
> every month of my data frame in this case based on volatility. For each
> month I want to know for each stock if it is in the most volatile quintile
> of if it is in one of the others.
> 
> So far I have come up with the following function (see below).
> Unfortunately, the function only works in some cases and often gives the
> following error:
> 
> Error in cut.default(df$VOLATILITY, unique(breaks), label = FALSE, na.rm
> =TRUE): 
> 
>  invalid number of intervals
> 
> Could you give me some advice on how to improve this code so that it works
> properly.
> 
> It's relatively urgent. Many thanks!
> 
> quintilesVolByMonth <- function(x){
> 
>  months<-as.vector(unique(x$DATE))  
> 
>  dfx<-data.frame()
> 
>  for(n in seq(1,length(months))){
> 
>    num<-5
> 
>    print(paste("Appending month",months[n],sep=""))
> 
>    df<-subset(x,DATE==months[n])
> 
>    breaks<-quantile(df$VOLATILITY,probs=seq(0,1, 1/num),na.rm=TRUE)
> 
>    df$volquintile <- cut(df$VOLATILITY,unique(breaks), 
> 
>                       label=FALSE, na.rm=TRUE)
> 
>    dfx<-rbind(dfx,df)
> 
>  }
> 
>  return(dfx)
> 
> }
> 
> Frame.Quintile <- quintilesVolByMonth(x) 
> 
> 
> 
> 
> 
> Example of data: (I have 30 years of monthly data for over 1000 stocks)
> Variable 'VolQuintile' is what I am trying to get.
> 
> 
> 
> DATE         STOCK     VOLATILITY     VolQuintile
> 
> 01/01/1990         STOCK 1               0.3
> 1
> 
> 01/01/1991         STOCK 2               0.15
> 4
> 
> 01/05/1992         STOCK 3               0.1
> 3
> 
> .
> 
> 
> 	[[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.
David Winsemius
Alameda, CA, USA