Choosing only a monthly subset using quantmod

问题

I'm using the latest 5 years of apple and google using the quant mode command

loadSymbols(c("AAPL", "GOOG"))
AAPL['2016::']

This is the dataset.

AAPL.Open AAPL.High AAPL.Low AAPL.Close AAPL.Volume AAPL.Adjusted
2016-01-04   25.6525   26.3425  25.5000    26.3375   270597600      24.36454
2016-01-05   26.4375   26.4625  25.6025    25.6775   223164000      23.75398
2016-01-06   25.1400   25.5925  24.9675    25.1750   273829600      23.28912
2016-01-07   24.6700   25.0325  24.1075    24.1125   324377600      22.30621
2016-01-08   24.6375   24.7775  24.1900    24.2400   283192000      22.42415

Can you please suggest a command I can use to choose the AAPL.Volume for a particular date of every month of every year?

Thanks.

回答1:

Try this with grep. This includes all rownames with the 4. month (April) in the AAPL data and the 5. column (AAPL.Volume).

AAPL[grep("-04-", row.names(as.data.frame(AAPL))),5]

First few rows:

head(AAPL[grep("-04-", row.names(as.data.frame(AAPL))),5])
           AAPL.Volume
2007-04-02   501992400
2007-04-03   583934400
2007-04-04   476784000
2007-04-05   355516000
2007-04-09   413341600
2007-04-10   352466800

EDIT
In your case, you might first want to discard the previous years:

AAPL <- AAPL['2016::']
head([grep("-04-", row.names(as.data.frame(AAPL))),5])
       AAPL.Volume
2016-04-01   103496000
2016-04-04   149424800
2016-04-05   106314800
2016-04-06   105616400
2016-04-07   127207600
2016-04-08    94326800

2. Edit
To pull out particular dates, modify the grep command like so:

head(AAPL[grep("-15$", row.names(as.data.frame(AAPL))),5])
           AAPL.Volume
2016-01-15   319335600
2016-03-15   160270800
2016-04-15   187756000
2016-06-15   117780800
2016-07-15   120548000
2016-08-15   103472800

回答2:

Based on the comment we want the volume on the 15th of the month for every month from April onwards in each year. Vo(AAPL) gives the volumes. Note that POSIXlt represents months starting from 0 so April is month 3. Note that some months may not have any data on the 15th.

Vo(AAPL)[with(as.POSIXlt(AAPL), mon >= 3 & mday == 15)]

If what is actually wanted is the first date that has data on or after the 15th for April onwards in the year then first extract the volumes on the 15 and after in each month on or after April, v.ok, and then find the index of the first one of those in each month, ix, and index by it.

v.ok <- Vo(AAPL)[with(as.POSIXlt(AAPL), mon >= 3 & mday >= 15)]
ix <- tapply(seq_along(v.ok), as.yearmon(time(v.ok)), head, 1)
v.ok[ix]

来源：https://stackoverflow.com/questions/66089610/choosing-only-a-monthly-subset-using-quantmod

标签

subset