thr3ads.net - R help - [R] snowfall [Jan 2011]

If this information is useful, please help other people find it:
Share via:

Santosh Srinivas

2011-Jan-12 08:47 UTC

[R] snowfall

Hello,

Just wondering why I am unable to run this in parallel.
A dput of my dataset is attached at the end. Please use to create my data
object.

I want to run this function in parallel (not sure if this is an efficient
implementation):

#Function to calculate the time to maturity for the option
require(fCalendar,quietly=TRUE) #Trying to calculate the trading days
require(fractalrock,quietly=TRUE) #Just to calculate the trading days
myFinCenter="Asia/Singapore"

getTimeToMaturity <- function(x){	
	tryCatch({
		toDt <- as.Date(as.character(x['EXPIRY_DT']), "%Y-%m-%d")
#Expiry Date
		fromDt <- as.Date(as.character(x['TIMESTAMP']),
"%Y-%m-%d")
#Trade Timestamp
		NoOfDays <- NROW(getTradingDates(toDt,fromDt))
		return(NoOfDays/252)
	},
	error = function (ex){
		#print (paste("Error in",toDt,fromDt))
		NoOfDays <- 0
		return(NoOfDays/252)
	}
	)
}


Question: The following two lines work but the third and parallel one
doesn't ... why?

1) > apply(dNiftyOpt,1,getTimeToMaturity) #Works
         1          2          3          4          5          6          7
8          9         10         11         12         13         14
15         16         17         18         19         20 
0.02380952 0.01984127 0.07936508 0.02380952 0.01984127 0.01190476 0.02777778
0.02380952 0.01984127 0.01190476 0.02380952 0.01984127 0.02380952 0.02380952
0.01984127 0.02380952 0.01984127 0.02380952 0.02380952 0.02777778


library(snowfall)
2) > sfInit()
snowfall 1.84 initialized: sequential execution, one CPU.
> sfApply(dNiftyOpt,1,getTimeToMaturity) #Works         1          2          3          4          5          6          7
8          9         10         11         12         13         14
15         16         17         18         19         20 
0.02380952 0.01984127 0.07936508 0.02380952 0.01984127 0.01190476 0.02777778
0.02380952 0.01984127 0.01190476 0.02380952 0.01984127 0.02380952 0.02380952
0.01984127 0.02380952 0.01984127 0.02380952 0.02380952 0.02777778 
> sfStop()

DOESN'T WORK: 3) > sfInit( parallel=TRUE, cpus=4 );
> sfApply(dNiftyOpt,1,getTimeToMaturity) #Added the time to maturity.DOESN'T WORK?
 1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 
 0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0
> sfStop();


My dataset:
dput(dNiftyOpt)

structure(list(INSTRUMENT = c("OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX", "OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX", "OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX"), SYMBOL = c("NIFTY",
"NIFTY", "NIFTY", "NIFTY",
"NIFTY", "NIFTY", "NIFTY", "NIFTY",
"NIFTY", "NIFTY", "NIFTY",
"NIFTY", "NIFTY", "NIFTY", "NIFTY",
"NIFTY", "NIFTY", "NIFTY",
"NIFTY", "NIFTY"), EXPIRY_DT = c("2004-01-29",
"2004-01-29",
"2004-01-29", "2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29", "2004-01-29"),
STRIKE_PR = c(1780,
1780, 1800, 1800, 1800, 1800, 1800, 1800, 1800, 1800, 1820, 1820, 
1820, 1830, 1830, 1830, 1830, 1840, 1840, 1850), OPTION_TYP = c("PE", 
"PE", "CE", "CE", "CE", "CE",
"PE", "PE", "PE", "PE", "CE",
"CE",
"PE", "CE", "CE", "PE", "PE",
"CE", "PE", "CE"), SETTLE_PR = c(27.4,
5.7, 152.95, 28.6, 70.45, 111.35, 14.75, 39.2, 8.6, 2.35, 20.4, 
54.2, 50.15, 18.35, 47.25, 51.75, 15.5, 14.95, 57.95, 26.3), 
    TIMESTAMP = c("2004-01-22", "2004-01-23",
"2004-01-02", "2004-01-22",
    "2004-01-23", "2004-01-27", "2004-01-21",
"2004-01-22", "2004-01-23",
    "2004-01-27", "2004-01-22", "2004-01-23",
"2004-01-22", "2004-01-22",
    "2004-01-23", "2004-01-22", "2004-01-23",
"2004-01-22", "2004-01-22",
    "2004-01-21"), Underlying = c(1770.5, 1847.55, 1946.05, 1770.5, 
    1847.55, 1904.7, 1824.6, 1770.5, 1847.55, 1904.7, 1770.5, 
    1847.55, 1770.5, 1770.5, 1847.55, 1770.5, 1847.55, 1770.5, 
    1770.5, 1824.6), UnderlyingVol = c(0.293906144944403, 0.331877179605752,

    0.129552369208600, 0.293906144944403, 0.331877179605752, 
    0.348918971622834, 0.276334860399362, 0.293906144944403, 
    0.331877179605752, 0.348918971622834, 0.293906144944403, 
    0.331877179605752, 0.293906144944403, 0.293906144944403, 
    0.331877179605752, 0.293906144944403, 0.331877179605752, 
    0.293906144944403, 0.293906144944403, 0.276334860399362)), .Names
c("INSTRUMENT",
"SYMBOL", "EXPIRY_DT", "STRIKE_PR",
"OPTION_TYP", "SETTLE_PR",
"TIMESTAMP", "Underlying", "UnderlyingVol"),
row.names = c("1",
"2", "3", "4", "5", "6",
"7", "8", "9", "10", "11",
"12", "13",
"14", "15", "16", "17", "18",
"19", "20"), class = "data.frame")

Uwe Ligges

2011-Jan-12 15:34 UTC

head link

[R] snowfall

You forgot to load the required packages on the client nodes by

  sfLibrary(fCalendar)
  sfLibrary(fractalrock)

and you really should not tryCatch without evaluating the errors for 
yourself.

Best wishes,
Uwe Ligges

On 12.01.2011 09:47, Santosh Srinivas wrote:> Hello,
>
> Just wondering why I am unable to run this in parallel.
> A dput of my dataset is attached at the end. Please use to create my data
> object.
>
> I want to run this function in parallel (not sure if this is an efficient
> implementation):
>
> #Function to calculate the time to maturity for the option
> require(fCalendar,quietly=TRUE) #Trying to calculate the trading days
> require(fractalrock,quietly=TRUE) #Just to calculate the trading days
> myFinCenter="Asia/Singapore"
>
> getTimeToMaturity<- function(x){	
> 	tryCatch({
> 		toDt<- as.Date(as.character(x['EXPIRY_DT']),
"%Y-%m-%d")
> #Expiry Date
> 		fromDt<- as.Date(as.character(x['TIMESTAMP']),
"%Y-%m-%d")
> #Trade Timestamp
> 		NoOfDays<- NROW(getTradingDates(toDt,fromDt))
> 		return(NoOfDays/252)
> 	},
> 	error = function (ex){
> 		#print (paste("Error in",toDt,fromDt))
> 		NoOfDays<- 0
> 		return(NoOfDays/252)
> 	}
> 	)
> }
>
>
> Question: The following two lines work but the third and parallel one
> doesn't ... why?
>
> 1)>  apply(dNiftyOpt,1,getTimeToMaturity) #Works
>           1          2          3          4          5          6         
7
> 8          9         10         11         12         13         14
> 15         16         17         18         19         20
> 0.02380952 0.01984127 0.07936508 0.02380952 0.01984127 0.01190476
0.02777778
> 0.02380952 0.01984127 0.01190476 0.02380952 0.01984127 0.02380952
0.02380952
> 0.01984127 0.02380952 0.01984127 0.02380952 0.02380952 0.02777778
>
>
> library(snowfall)
> 2)>  sfInit()
> snowfall 1.84 initialized: sequential execution, one CPU.
>
>> sfApply(dNiftyOpt,1,getTimeToMaturity) #Works
>           1          2          3          4          5          6         
7
> 8          9         10         11         12         13         14
> 15         16         17         18         19         20
> 0.02380952 0.01984127 0.07936508 0.02380952 0.01984127 0.01190476
0.02777778
> 0.02380952 0.01984127 0.01190476 0.02380952 0.01984127 0.02380952
0.02380952
> 0.01984127 0.02380952 0.01984127 0.02380952 0.02380952 0.02777778
>
>> sfStop()
>
>
> DOESN'T WORK: 3)
>> sfInit( parallel=TRUE, cpus=4 );
>> sfApply(dNiftyOpt,1,getTimeToMaturity) #Added the time to maturity.
> DOESN'T WORK?
>   1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20
>   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0
>> sfStop();
>
>
>
> My dataset:
> dput(dNiftyOpt)
>
> structure(list(INSTRUMENT = c("OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX",
> "OPTIDX", "OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX", "OPTIDX", "OPTIDX",
> "OPTIDX", "OPTIDX", "OPTIDX",
"OPTIDX", "OPTIDX", "OPTIDX", "OPTIDX",
> "OPTIDX", "OPTIDX"), SYMBOL = c("NIFTY",
"NIFTY", "NIFTY", "NIFTY",
> "NIFTY", "NIFTY", "NIFTY", "NIFTY",
"NIFTY", "NIFTY", "NIFTY",
> "NIFTY", "NIFTY", "NIFTY", "NIFTY",
"NIFTY", "NIFTY", "NIFTY",
> "NIFTY", "NIFTY"), EXPIRY_DT =
c("2004-01-29", "2004-01-29",
> "2004-01-29", "2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29",
> "2004-01-29", "2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29",
> "2004-01-29", "2004-01-29", "2004-01-29",
"2004-01-29", "2004-01-29",
> "2004-01-29", "2004-01-29", "2004-01-29"),
STRIKE_PR = c(1780,
> 1780, 1800, 1800, 1800, 1800, 1800, 1800, 1800, 1800, 1820, 1820,
> 1820, 1830, 1830, 1830, 1830, 1840, 1840, 1850), OPTION_TYP =
c("PE",
> "PE", "CE", "CE", "CE",
"CE", "PE", "PE", "PE", "PE",
"CE", "CE",
> "PE", "CE", "CE", "PE",
"PE", "CE", "PE", "CE"), SETTLE_PR =
c(27.4,
> 5.7, 152.95, 28.6, 70.45, 111.35, 14.75, 39.2, 8.6, 2.35, 20.4,
> 54.2, 50.15, 18.35, 47.25, 51.75, 15.5, 14.95, 57.95, 26.3),
>      TIMESTAMP = c("2004-01-22", "2004-01-23",
"2004-01-02", "2004-01-22",
>      "2004-01-23", "2004-01-27",
"2004-01-21", "2004-01-22", "2004-01-23",
>      "2004-01-27", "2004-01-22",
"2004-01-23", "2004-01-22", "2004-01-22",
>      "2004-01-23", "2004-01-22",
"2004-01-23", "2004-01-22", "2004-01-22",
>      "2004-01-21"), Underlying = c(1770.5, 1847.55, 1946.05,
1770.5,
>      1847.55, 1904.7, 1824.6, 1770.5, 1847.55, 1904.7, 1770.5,
>      1847.55, 1770.5, 1770.5, 1847.55, 1770.5, 1847.55, 1770.5,
>      1770.5, 1824.6), UnderlyingVol = c(0.293906144944403,
0.331877179605752,
>
>      0.129552369208600, 0.293906144944403, 0.331877179605752,
>      0.348918971622834, 0.276334860399362, 0.293906144944403,
>      0.331877179605752, 0.348918971622834, 0.293906144944403,
>      0.331877179605752, 0.293906144944403, 0.293906144944403,
>      0.331877179605752, 0.293906144944403, 0.331877179605752,
>      0.293906144944403, 0.293906144944403, 0.276334860399362)), .Names >
c("INSTRUMENT",
> "SYMBOL", "EXPIRY_DT", "STRIKE_PR",
"OPTION_TYP", "SETTLE_PR",
> "TIMESTAMP", "Underlying", "UnderlyingVol"),
row.names = c("1",
> "2", "3", "4", "5", "6",
"7", "8", "9", "10", "11",
"12", "13",
> "14", "15", "16", "17",
"18", "19", "20"), class = "data.frame")
>
> ______________________________________________
> 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.

Reasonably Related Threads

Search for more reasonably related threads

R help - Jan 2011 - snowfall

[R] snowfall

[R] snowfall

Reasonably Related Threads