CODEWITHSUNDEEP
rdata.table
Ricky Bobby
Ricky Bobby

Reputation: 7608

Calculate marketshare per day in R

I'm currently working on a data.table (myDT) with the folowing 3 columns :

TRADE_DATE TRADER VOLUME
01-MAY-2013  T1  100
01-MAY-2013  T2  200
01-MAY-2013  T3  500
01-MAY-2013  T4  200
02-MAY-2013  T1  400
02-MAY-2013  T2  500
02-MAY-2013  T3  50
02-MAY-2013  T4  50

I would like to find a way to calculate a share volume per trader per day. Currently I'm calculating the total volume per day then merge with the previous table to calculate the market share of each trader per day. Is there a more straightforward way to do it with data.tables ?

I attached herunder my code :

  daylyVolume<-myDT[,list(DAILY_VOLUME=sum(VOLUME)),by="TRADE_DATE"]
  myDT<-merge(myDT,daylyVolume,all=TRUE,by='TRADE_DATE')
  myDT$"SHARE_VOLUME"<-100*myDT$"VOLUME"/myDT$"DAILY_VOLUME"

The result is the following:

TRADE_DATE TRADER VOLUME SHARE  DAYLY_VOLUME
01-MAY-2013  T1  100      10       1000
01-MAY-2013  T2  200      20       1000
01-MAY-2013  T3  500      50       1000
01-MAY-2013  T4  200      20       1000
02-MAY-2013  T1  400      40       1000
02-MAY-2013  T2  500      50       1000
02-MAY-2013  T3  50       5        1000
02-MAY-2013  T4  50       5        1000

Upvotes: 2

Views: 1622

Answers (2)

Jonas Tundo
Jonas Tundo

Reputation: 6197

Do you want this?

myDT[,list(share= VOLUME/sum(VOLUME)*100),by="TRADE_DATE"]

Upvotes: 1

juba
juba

Reputation: 49033

A simple way is to do this :

dt[,list(VOLUME,DAILY_VOLUME=sum(VOLUME),SHARE=VOLUME/sum(VOLUME)*100),by="TRADE_DATE"]

    TRADE_DATE VOLUME DAILY_VOLUME SHARE
1: 01-MAY-2013    100         1000    10
2: 01-MAY-2013    200         1000    20
3: 01-MAY-2013    500         1000    50
4: 01-MAY-2013    200         1000    20
5: 02-MAY-2013    400         1000    40
6: 02-MAY-2013    500         1000    50
7: 02-MAY-2013     50         1000     5
8: 02-MAY-2013     50         1000     5

A more efficient way, if you just want to compute the share, is :

R> dt[,SHARE:=VOLUME/sum(VOLUME)*100,by="TRADE_DATE"]
R> dt

    TRADE_DATE TRADER VOLUME SHARE
1: 01-MAY-2013     T1    100    10
2: 01-MAY-2013     T2    200    20
3: 01-MAY-2013     T3    500    50
4: 01-MAY-2013     T4    200    20
5: 02-MAY-2013     T1    400    40
6: 02-MAY-2013     T2    500    50
7: 02-MAY-2013     T3     50     5
8: 02-MAY-2013     T4     50     5

Upvotes: 2

Related Questions