Re: [R] Better way of writing R code
Dear Sir, Thanks a lot for your great help. Do appreciate it a lot. In my earlier mail, where I had attached some files, I have realized yesterday that instead of sending the R code customized by me based on your guidance, I had by mistake attached the contents of email. I do apologize to you for the same. Thanks once again and sorry for the inconvenience caused by me. Regards Katherine --- On Fri, 5/4/13, Adams, Jean wrote: From: Adams, Jean Subject: Re: [R] Better way of writing R code To: "Katherine Gobin" Cc: "R help" Date: Friday, 5 April, 2013, 2:40 PM Katherine, To preserve the original order, you could create a new variable for the currency data frame (BEFORE the merges), then use this variable to reorder at the end. currency_df$orig.order <- 1:dim(currency_df)[1] You can do another merge for the other currency, you just need to specify the columns that you want to merge by. The rate information will be called rate.x for the first currency (from the first merge) and rate.y for the other currency (from the second merge). both2 <- merge(both, rate_df, by.x=c("other_currency", "tenor"), by.y=c("currency", "tenor"), all.x=TRUE) Then reorder. both2 <- both2[order(both2$orig.order), ] Jean On Thu, Apr 4, 2013 at 3:19 AM, Katherine Gobin wrote: Dear Mr Adams, I sincerely apologize for taking the liberty of writing to you. I wholeheartedly thank you for the wonderful solution you had provided me yesterday. I have customized the R code you had provided and it's yielding the results. I can't imagine me repeating the 1 lines code after receving such a powerful solution from you. In future it will save lots of efforts from my side as I always deal with such situation. There is one small problem though - I am dealing with pair of currencies e.g. currency other_currency transaction USD EURO Buy USD CAD Buy GBP CHF Sell SEK USD Buy The R code gives me the currency rates (w.r.t. appropriate "tenor"), however, I need the corresponding rates pertaining to the other currency too i.e. in the first case, the maturity period applicable is one month so the R - code gives me one month LIBOR wr.t. USD, but I need the corresponding one month LIBOR w.r.t. the other currency i.e. EURO in this case. I tried to improve upon the merge statement and used "?merge", but couldn't. Another problem is the order of the original portfolio is not mainteained , but I think I can manage the order. With warm regards Katherine --- On Wed, 3/4/13, Adams, Jean wrote: From: Adams, Jean Subject: Re: [R] Better way of writing R code To: "Katherine Gobin" Cc: "R help" Date: Wednesday, 3 April, 2013, 2:08 PM Katherine, You don't need to convert rate_df into tabular form. You just need to categorize each row in currency_df into a "tenor". Then you can merge the two data frames (by currency and tenor). For example ... # convert dates to R dates, to calculate the number of days to maturity# I am assuming this is the number of days from the current date to the maturity date currency_df$maturity <- as.Date(currency_df$maturity_date, "%d/%m/%Y")currency_df$current <- as.Date(currency_df$current_date, "%d/%m/%Y")currency_df$days2mature <- as.numeric(currency_df$maturity - currency_df$current) # categorize the number of days to maturity as you wish# you may need to change the breaks= option to suit your needs# read about the cut function to make sure you get the cut points included in the proper category, ?cut currency_df$tenor <- cut(currency_df$days2mature, breaks=c(0, 1, 7, 14, seq(from=30.5, length=12, by=30.5)),labels=c("1 day", "1 week", "2 weeks", "1 month", paste(2:12, "months"))) # merge the currency_df and rate_df# this will work better with real data, since the example data you provided didn't have matching tenorsboth <- merge(currency_df, rate_df, all.x=TRUE) Jean On Wed, Apr 3, 2013 at 5:21 AM, Katherine Gobin wrote: Dear R forum, (Pl note this is not a finance problem) I have two data.frames as currency_df = data.frame(current_date = c("3/4/2013", "3/4/2013", "3/4/2013", "3/4/2013"), issue_date = c("27/11/2012", "9/12/2012", "14/01/2013", "28/02/2013"), maturity_date = c("27/04/2013", "3/5/2013", "14/6/2013", "28/06/2013"), currency = c("USD", "USD", "GBP", "SEK"), other_currency = c("EURO", "CAD", "CHF", "USD"), transactio
Re: [R] Better way of writing R code
On Apr 4, 2013, at 6:48 AM, Gabor Grothendieck wrote: > On Thu, Apr 4, 2013 at 9:32 AM, Adams, Jean wrote: >> Katherine, >> >> You should cc the R-help on all correspondence. >> The more eyes that see your query, the quicker and probably the better the >> response will be. >> Send your message as plain text with no attachments ... so, include your >> code, and use dput() to share some example data. >> > > Although many types of attachments are not allowed it seems that .txt, > .R, .png, .pdf and possibly certain other types are accepted. I think it varies with the mail-client and how it labels the type of each attachment. At one time I read that "text" files needed to labeled as MIME-text to pass scrutiny, so it's possible that some mailers will send .R files in an acceptable fashion but not others. I have tried to find where this is documented but have failed. It may have been one of the changes to the Posting Guide that occurred a couple of years ago. I know from experience that most attempts to send c-s-v files with a .csv extension will fail, but that when sent with .txt extensions will succeed. I think .jpeg files (and perhaps .ps) may make it through intact. -- David Winsemius Alameda, CA, USA __ 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.
Re: [R] Better way of writing R code
Hi, You can preserve the order using join(). Though, in this case, merge would be easier. library(plyr) both<- join(currency_df,rate_df,type="left") names(both)[14:15]<- paste0("currency_",names(both)[14:15]) names(rate_df)[2]<- "other_currency" both2<- join(both,rate_df,type="left",by=c("other_currency","tenor")) names(both2)[16:17]<- paste0("other_currency_",names(both2)[16:17]) A.K. - Original Message - From: "Adams, Jean" To: Katherine Gobin Cc: R help Sent: Friday, April 5, 2013 9:40 AM Subject: Re: [R] Better way of writing R code Katherine, To preserve the original order, you could create a new variable for the currency data frame (BEFORE the merges), then use this variable to reorder at the end. currency_df$orig.order <- 1:dim(currency_df)[1] You can do another merge for the other currency, you just need to specify the columns that you want to merge by. The rate information will be called rate.x for the first currency (from the first merge) and rate.y for the other currency (from the second merge). both2 <- merge(both, rate_df, by.x=c("other_currency", "tenor"), by.y=c("currency", "tenor"), all.x=TRUE) Then reorder. both2 <- both2[order(both2$orig.order), ] Jean On Thu, Apr 4, 2013 at 3:19 AM, Katherine Gobin wrote: > Dear Mr Adams, > > I sincerely apologize for taking the liberty of writing to you. I > wholeheartedly thank you for the wonderful solution you had provided me > yesterday. I have customized the R code you had provided and it's yielding > the results. I can't imagine me repeating the 1 lines code after > receving such a powerful solution from you. In future it will save lots of > efforts from my side as I always deal with such situation. > > There is one small problem though - > > I am dealing with pair of currencies > > e.g. currency other_currency transaction > USD EURO Buy > USD CAD Buy > GBP CHF Sell > SEK USD Buy > > > The R code gives me the currency rates (w.r.t. appropriate "tenor"), > however, I need the corresponding rates pertaining to the other currency > too i.e. in the first case, the maturity period applicable is one month so > the R - code gives me one month LIBOR wr.t. USD, but I need the > corresponding one month LIBOR w.r.t. the other currency i.e. EURO in this > case. > > I tried to improve upon the merge statement and used "?merge", but > couldn't. Another problem is the order of the original portfolio is not > mainteained , but I think I can manage the order. > > With warm regards > > > Katherine > > > > > > > > > --- On *Wed, 3/4/13, Adams, Jean * wrote: > > > From: Adams, Jean > Subject: Re: [R] Better way of writing R code > To: "Katherine Gobin" > Cc: "R help" > Date: Wednesday, 3 April, 2013, 2:08 PM > > Katherine, > > You don't need to convert rate_df into tabular form. You just need to > categorize each row in currency_df into a "tenor". Then you can merge the > two data frames (by currency and tenor). For example ... > > # convert dates to R dates, to calculate the number of days to maturity > # I am assuming this is the number of days from the current date to the > maturity date > currency_df$maturity <- as.Date(currency_df$maturity_date, "%d/%m/%Y") > currency_df$current <- as.Date(currency_df$current_date, "%d/%m/%Y") > currency_df$days2mature <- as.numeric(currency_df$maturity - > currency_df$current) > > # categorize the number of days to maturity as you wish > # you may need to change the breaks= option to suit your needs > # read about the cut function to make sure you get the cut points included > in the proper category, ?cut > currency_df$tenor <- cut(currency_df$days2mature, breaks=c(0, 1, 7, 14, > seq(from=30.5, length=12, by=30.5)), > labels=c("1 day", "1 week", "2 weeks", "1 month", paste(2:12, "months"))) > > # merge the currency_df and rate_df > # this will work better with real data, since the example data you > provided didn't have matching tenors > both <- merge(currency_df, rate_df, all.x=TRUE) > > Jean > > > > On Wed, Apr 3, 2013 at 5:21 AM, Katherine Gobin > http://mc/compose?to=katherine_go...@yahoo.com> > > wrote: > > Dear R forum, > > (Pl note this is not a finance problem) > > I have two data.frames as > > currency_df = data.frame(current_date = c("3/4/2013&q
Re: [R] Better way of writing R code
Katherine, To preserve the original order, you could create a new variable for the currency data frame (BEFORE the merges), then use this variable to reorder at the end. currency_df$orig.order <- 1:dim(currency_df)[1] You can do another merge for the other currency, you just need to specify the columns that you want to merge by. The rate information will be called rate.x for the first currency (from the first merge) and rate.y for the other currency (from the second merge). both2 <- merge(both, rate_df, by.x=c("other_currency", "tenor"), by.y=c("currency", "tenor"), all.x=TRUE) Then reorder. both2 <- both2[order(both2$orig.order), ] Jean On Thu, Apr 4, 2013 at 3:19 AM, Katherine Gobin wrote: > Dear Mr Adams, > > I sincerely apologize for taking the liberty of writing to you. I > wholeheartedly thank you for the wonderful solution you had provided me > yesterday. I have customized the R code you had provided and it's yielding > the results. I can't imagine me repeating the 1 lines code after > receving such a powerful solution from you. In future it will save lots of > efforts from my side as I always deal with such situation. > > There is one small problem though - > > I am dealing with pair of currencies > > e.g. currencyother_currencytransaction > USD EURO Buy > USD CAD Buy > GBP CHF Sell > SEK USD Buy > > > The R code gives me the currency rates (w.r.t. appropriate "tenor"), > however, I need the corresponding rates pertaining to the other currency > too i.e. in the first case, the maturity period applicable is one month so > the R - code gives me one month LIBOR wr.t. USD, but I need the > corresponding one month LIBOR w.r.t. the other currency i.e. EURO in this > case. > > I tried to improve upon the merge statement and used "?merge", but > couldn't. Another problem is the order of the original portfolio is not > mainteained , but I think I can manage the order. > > With warm regards > > > Katherine > > > > > > > > > --- On *Wed, 3/4/13, Adams, Jean * wrote: > > > From: Adams, Jean > Subject: Re: [R] Better way of writing R code > To: "Katherine Gobin" > Cc: "R help" > Date: Wednesday, 3 April, 2013, 2:08 PM > > Katherine, > > You don't need to convert rate_df into tabular form. You just need to > categorize each row in currency_df into a "tenor". Then you can merge the > two data frames (by currency and tenor). For example ... > > # convert dates to R dates, to calculate the number of days to maturity > # I am assuming this is the number of days from the current date to the > maturity date > currency_df$maturity <- as.Date(currency_df$maturity_date, "%d/%m/%Y") > currency_df$current <- as.Date(currency_df$current_date, "%d/%m/%Y") > currency_df$days2mature <- as.numeric(currency_df$maturity - > currency_df$current) > > # categorize the number of days to maturity as you wish > # you may need to change the breaks= option to suit your needs > # read about the cut function to make sure you get the cut points included > in the proper category, ?cut > currency_df$tenor <- cut(currency_df$days2mature, breaks=c(0, 1, 7, 14, > seq(from=30.5, length=12, by=30.5)), > labels=c("1 day", "1 week", "2 weeks", "1 month", paste(2:12, "months"))) > > # merge the currency_df and rate_df > # this will work better with real data, since the example data you > provided didn't have matching tenors > both <- merge(currency_df, rate_df, all.x=TRUE) > > Jean > > > > On Wed, Apr 3, 2013 at 5:21 AM, Katherine Gobin > http://mc/compose?to=katherine_go...@yahoo.com> > > wrote: > > Dear R forum, > > (Pl note this is not a finance problem) > > I have two data.frames as > > currency_df = data.frame(current_date = c("3/4/2013", "3/4/2013", > "3/4/2013", "3/4/2013"), issue_date = c("27/11/2012", "9/12/2012", > "14/01/2013", "28/02/2013"), maturity_date = c("27/04/2013", "3/5/2013", > "14/6/2013", "28/06/2013"), currency = c("USD", "USD", "GBP", "SEK"), > other_currency = c("EURO", "CAD", "CHF", "USD"), transaction = c("Buy", > "Buy", "Sell", "Buy"), units_currency = c(10, 25000, 15, 4), > units_other_currency = c(
Re: [R] Better way of writing R code
Dear Sirs, I sincerely apologize for the blunder at my end. Problem is I was told that one cannot or should not send any ATTACHMENTS. In the past, when I had tried to attach some files and the message was displayed less the attachment. Also, at times it becomes very difficult to attach the csv file. As my input files contain the csv files and since I was under the impression that we cannot attach the files to this forum. I once again apologize to all of you for the inconvenience caused. Regards Katherine --- On Thu, 4/4/13, Gabor Grothendieck wrote: From: Gabor Grothendieck Subject: Re: [R] Better way of writing R code To: "Adams, Jean" Cc: "Katherine Gobin" , "R help" Date: Thursday, 4 April, 2013, 2:48 PM On Thu, Apr 4, 2013 at 9:32 AM, Adams, Jean wrote: > Katherine, > > You should cc the R-help on all correspondence. > The more eyes that see your query, the quicker and probably the better the > response will be. > Send your message as plain text with no attachments ... so, include your > code, and use dput() to share some example data. > Although many types of attachments are not allowed it seems that .txt, .R, .png, .pdf and possibly certain other types are accepted. [[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.
Re: [R] Better way of writing R code
On Thu, Apr 4, 2013 at 9:32 AM, Adams, Jean wrote: > Katherine, > > You should cc the R-help on all correspondence. > The more eyes that see your query, the quicker and probably the better the > response will be. > Send your message as plain text with no attachments ... so, include your > code, and use dput() to share some example data. > Although many types of attachments are not allowed it seems that .txt, .R, .png, .pdf and possibly certain other types are accepted. __ 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.
Re: [R] Better way of writing R code
Katherine, You should cc the R-help on all correspondence. The more eyes that see your query, the quicker and probably the better the response will be. Send your message as plain text with no attachments ... so, include your code, and use dput() to share some example data. Jean On Thu, Apr 4, 2013 at 3:19 AM, Katherine Gobin wrote: > Dear Mr Adams, > > I sincerely apologize for taking the liberty of writing to you. I > wholeheartedly thank you for the wonderful solution you had provided me > yesterday. I have customized the R code you had provided and it's yielding > the results. I can't imagine me repeating the 1 lines code after > receving such a powerful solution from you. In future it will save lots of > efforts from my side as I always deal with such situation. > > There is one small problem though - > > I am dealing with pair of currencies > > e.g. currencyother_currencytransaction > USD EURO Buy > USD CAD Buy > GBP CHF Sell > SEK USD Buy > > > The R code gives me the currency rates (w.r.t. appropriate "tenor"), > however, I need the corresponding rates pertaining to the other currency > too i.e. in the first case, the maturity period applicable is one month so > the R - code gives me one month LIBOR wr.t. USD, but I need the > corresponding one month LIBOR w.r.t. the other currency i.e. EURO in this > case. > > I tried to improve upon the merge statement and used "?merge", but > couldn't. Another problem is the order of the original portfolio is not > mainteained , but I think I can manage the order. > > With warm regards > > > Katherine > > > > > > > > > --- On *Wed, 3/4/13, Adams, Jean * wrote: > > > From: Adams, Jean > Subject: Re: [R] Better way of writing R code > To: "Katherine Gobin" > Cc: "R help" > Date: Wednesday, 3 April, 2013, 2:08 PM > > Katherine, > > You don't need to convert rate_df into tabular form. You just need to > categorize each row in currency_df into a "tenor". Then you can merge the > two data frames (by currency and tenor). For example ... > > # convert dates to R dates, to calculate the number of days to maturity > # I am assuming this is the number of days from the current date to the > maturity date > currency_df$maturity <- as.Date(currency_df$maturity_date, "%d/%m/%Y") > currency_df$current <- as.Date(currency_df$current_date, "%d/%m/%Y") > currency_df$days2mature <- as.numeric(currency_df$maturity - > currency_df$current) > > # categorize the number of days to maturity as you wish > # you may need to change the breaks= option to suit your needs > # read about the cut function to make sure you get the cut points included > in the proper category, ?cut > currency_df$tenor <- cut(currency_df$days2mature, breaks=c(0, 1, 7, 14, > seq(from=30.5, length=12, by=30.5)), > labels=c("1 day", "1 week", "2 weeks", "1 month", paste(2:12, "months"))) > > # merge the currency_df and rate_df > # this will work better with real data, since the example data you > provided didn't have matching tenors > both <- merge(currency_df, rate_df, all.x=TRUE) > > Jean > > > > On Wed, Apr 3, 2013 at 5:21 AM, Katherine Gobin > http://mc/compose?to=katherine_go...@yahoo.com> > > wrote: > > Dear R forum, > > (Pl note this is not a finance problem) > > I have two data.frames as > > currency_df = data.frame(current_date = c("3/4/2013", "3/4/2013", > "3/4/2013", "3/4/2013"), issue_date = c("27/11/2012", "9/12/2012", > "14/01/2013", "28/02/2013"), maturity_date = c("27/04/2013", "3/5/2013", > "14/6/2013", "28/06/2013"), currency = c("USD", "USD", "GBP", "SEK"), > other_currency = c("EURO", "CAD", "CHF", "USD"), transaction = c("Buy", > "Buy", "Sell", "Buy"), units_currency = c(10, 25000, 15, 4), > units_other_currency = c(78000, 25350, 99200, 6150)) > > rate_df = > data.frame(date = > c("28/3/2013","27/3/2013","26/3/2013","25/3/2013","28/3/2013","27/3/2013","26/3/2013", > "25/3/2013","28/3/2013","27/3/2013","26/3/2013","25/3/2013","28/3/2013","27/3/2013","26/3/2013", > > "
Re: [R] Better way of writing R code
Katherine, You don't need to convert rate_df into tabular form. You just need to categorize each row in currency_df into a "tenor". Then you can merge the two data frames (by currency and tenor). For example ... # convert dates to R dates, to calculate the number of days to maturity # I am assuming this is the number of days from the current date to the maturity date currency_df$maturity <- as.Date(currency_df$maturity_date, "%d/%m/%Y") currency_df$current <- as.Date(currency_df$current_date, "%d/%m/%Y") currency_df$days2mature <- as.numeric(currency_df$maturity - currency_df$current) # categorize the number of days to maturity as you wish # you may need to change the breaks= option to suit your needs # read about the cut function to make sure you get the cut points included in the proper category, ?cut currency_df$tenor <- cut(currency_df$days2mature, breaks=c(0, 1, 7, 14, seq(from=30.5, length=12, by=30.5)), labels=c("1 day", "1 week", "2 weeks", "1 month", paste(2:12, "months"))) # merge the currency_df and rate_df # this will work better with real data, since the example data you provided didn't have matching tenors both <- merge(currency_df, rate_df, all.x=TRUE) Jean On Wed, Apr 3, 2013 at 5:21 AM, Katherine Gobin wrote: > Dear R forum, > > (Pl note this is not a finance problem) > > I have two data.frames as > > currency_df = data.frame(current_date = c("3/4/2013", "3/4/2013", > "3/4/2013", "3/4/2013"), issue_date = c("27/11/2012", "9/12/2012", > "14/01/2013", "28/02/2013"), maturity_date = c("27/04/2013", "3/5/2013", > "14/6/2013", "28/06/2013"), currency = c("USD", "USD", "GBP", "SEK"), > other_currency = c("EURO", "CAD", "CHF", "USD"), transaction = c("Buy", > "Buy", "Sell", "Buy"), units_currency = c(10, 25000, 15, 4), > units_other_currency = c(78000, 25350, 99200, 6150)) > > rate_df = > data.frame(date = > c("28/3/2013","27/3/2013","26/3/2013","25/3/2013","28/3/2013","27/3/2013","26/3/2013", > "25/3/2013","28/3/2013","27/3/2013","26/3/2013","25/3/2013","28/3/2013","27/3/2013","26/3/2013", > > "25/3/2013","28/3/2013","27/3/2013","26/3/2013","25/3/2013","28/3/2013","27/3/2013","26/3/2013", > "25/3/2013","28/3/2013","27/3/2013","26/3/2013","25/3/2013","28/3/2013","27/3/2013","26/3/2013", > "25/3/2013","28/3/2013","27/3/2013","26/3/2013","25/3/2013"), > > currency = c("USD","USD","USD","USD", "USD", "USD", > "USD","USD","USD","USD", "USD","USD", > "GBP","GBP","GBP","GBP","GBP","GBP","GBP","GBP", "GBP","GBP", "GBP","GBP", > "EURO","EURO","EURO","EURO","EURO","EURO","EURO", "EURO", "EURO","EURO", > "EURO","EURO"), > > tenor = c("1 day","1 day","1 day","1 day","1 week","1 week","1 week","1 > week","2 weeks","2 weeks","2 weeks","2 weeks","1 day","1 day","1 day","1 > day","1 week","1 week","1 week","1 week","2 weeks","2 weeks","2 weeks","2 > weeks","1 day","1 day","1 day","1 day","1 week","1 week","1 week","1 > week","2 weeks","2 weeks","2 weeks","2 weeks"), > > rate = > c(0.156,0.157,0.157,0.155,0.1752,0.1752,0.1752,0.1752,0.1752,0.1752,0.1752, > 0.1752,0.48625, > 0.485,0.48625,0.4825,0.49,0.49125,0.4925,0.49,0.49375,0.49125,0.4925, > 0.49125,0.02643,0.02214, > 0.02214,0.01929,0.034,0.034,0.034125,0.034,0.044,0.044, 0.041,0.045)) > > # ___ > > # 1st data.frame > > > currency_df > current_date issue_date maturity_date currency > 1 3/4/2013 27/11/201227/04/2013 USD > 2 3/4/2013 9/12/2012 3/5/2013 USD > 3 3/4/2013 14/01/2013 14/6/2013 GBP > 4 3/4/2013 28/02/201328/06/2013 SEK > other_currency transaction units_currency > 1 > EURO Buy 10 > 2CAD Buy 25000 > 3CHFSell 15 > 4USD Buy 4 > units_other_currency > 178000 > 2 > 25350 > 399200 > 4 6150 > > # > ... > > # 2nd data.frame > > > rate_df > date currency tenor rate > 1 28/3/2013 USD 1 day 0.156000 > 2 27/3/2013 USD 1 day 0.157000 > 3 26/3/2013 USD 1 day 0.157000 > 4 25/3/2013 USD 1 day 0.155000 > 5 28/3/2013 USD 1 week 0.175200 > 6 27/3/2013 USD 1 week > 0.175200 > 7 26/3/2013 USD 1 week 0.175200 > 8 25/3/2013 USD 1 week 0.175200 > 9 28/3/2013 USD 2 weeks 0.175200 > 10 27/3/2013 USD 2 weeks 0.175200 > 11 26/3/2013 USD 2 weeks 0.175200 > 12 25/3/2013 USD 2 weeks 0.175200 > 13 28/3/2013 GBP 1 day 0.486250 > 14 27/3/2013 GBP 1 day 0.485000 > 15 26/3/2013 GBP 1 day 0.486250 > 16 25/3/2013 GBP 1 day 0.482500 > 17 28/3/2013 GBP 1 week 0.49 > 18 27/3/2013 GBP 1 week 0.491250 > 19 26/3/2013 GBP 1 week 0.492500 > 20 > 25/3/2013 GBP 1 week 0.49 > 21 28/3/2013 GBP 2 weeks 0.493750 > 22 27/3/2013 GBP 2 weeks 0.491250 > 2