On Thu, 05 Sep 2024, Brian G. Peterson via R-SIG-Finance writes:
> the 'returns' package has just turned them into whole numbers rather
> than decimals. The numbers all agree.
Thank you, Brian. Indeed, I do hope that the numbers
are the same ;-)
Amarjit, it's just that PMwR::returns _prints_ the
returns as percentages. To give an example:
library("quantmod")
S <- getSymbols("SPY", src = "yahoo",
from = as.Date("2020-01-01"),
return.class = "zoo",
periodicity = "daily",
auto.assign = FALSE)[, 6]
library("PMwR")
returns(S, period = "month")
## Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec YTD
## 2020 -1.0 -7.9 -12.5 12.7 4.8 1.8 5.9 7.0 -3.7 -2.5 10.9 3.7 17.2
## 2021 -1.0 2.8 4.5 5.3 0.7 2.2 2.4 3.0 -4.7 7.0 -0.8 4.6 28.7
## 2022 -5.3 -3.0 3.8 -8.8 0.2 -8.2 9.2 -4.1 -9.2 8.1 5.6 -5.8 -18.2
## 2023 6.3 -2.5 3.7 1.6 0.5 6.5 3.3 -1.6 -4.7 -2.2 9.1 4.6 26.2
## 2024 1.6 5.2 3.3 -4.0 5.1 3.5 1.2 2.3 -2.3 16.6
To see the raw numbers, a convenient way is to use
'as.matrix', which drops the class and leaves a numeric
matrix (with options(digits = 3)):
as.matrix(returns(S, period = "month"))
## 1 2 3 4 5 6
## 2020 -0.00967 -0.0792 -0.1249 0.1270 0.04765 0.0177
## 2021 -0.01019 0.0278 0.0454 0.0529 0.00657 0.0224
## 2022 -0.05274 -0.0295 0.0376 -0.0878 0.00226 -0.0825
## 2023 0.06289 -0.0251 0.0371 0.0160 0.00462 0.0648
## 2024 0.01593 0.0522 0.0327 -0.0403 0.05058 0.0353
## 7 8 9 10 11 12 YTD
## 2020 0.0589 0.0698 -0.0374 -0.0249 0.10878 0.0370 0.172
## 2021 0.0244 0.0298 -0.0466 0.0702 -0.00803 0.0462 0.287
## 2022 0.0921 -0.0408 -0.0924 0.0813 0.05559 -0.0576 -0.182
## 2023 0.0327 -0.0163 -0.0474 -0.0217 0.09134 0.0457 0.262
## 2024 0.0121 0.0234 -0.0226 NA NA NA 0.166
kind regards
Enrico
>On Thu, 2024-09-05 at 12:35 +0100, Amarjit Chandhial via R-SIG-Finance
>wrote:
>>
>>
>> Thanks Enrico!
>>
>> If try on SPY for the year 2023
>>
>> > calendarReturnTable(spyRets, digits = 4, percent = FALSE)
>> Using 'Value' as value column. Use 'value.var' to override
>> Jan Feb Mar Apr May Jun Jul Aug
>> Sep
>> Oct Nov Dec Annual
>> 0.0674 -0.0251 0.0371 0.0160 0.0046 0.0648 0.0327 -0.0163 -
>> 0.0474
>> -0.0217 0.0913 0.0457 0.2671
>> DD
>> 0.0997
>> > PMwR::returns(SPY, period = "month")
>> Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec YTD
>> 2023 6.7 -2.5 3.7 1.6 0.5 6.5 3.3 -1.6 -4.7 -2.2 9.1 4.6 26.7 >
>> PerformanceAnalytics::table.AnnualizedReturns(spyRets, scale = 249,
>> geometric = TRUE)
>> Adjusted
>> Annualized Return 0.2671
>> Annualized Std Dev 0.1301
>> Annualized Sharpe (Rf=0%) 2.0525
>> > PerformanceAnalytics::maxDrawdown(spyRets)
>> [1] 0.09974311
>>
>>
>> Amarjit
>>
>>
>>
>> ------ Original Message ------
>> From: e...@enricoschumann.net
>> To: a.chandh...@btinternet.com Cc: r-sig-finance@r-project.org
>> Sent: Thursday, September 5th 2024, 10:10
>> Subject: Re: [R-SIG-Finance]
>> PerformanceAnalytics::table.CalendarReturns
>> On Wed, 04 Sep 2024, Amarjit Chandhial via R-SIG-Finance writes: >
>> Hi, >
>> > Are there any plans for >
>> https://timelyportfolio.github.io/PerformanceAnalytics/reference/table.CalendarReturns.html
>>
>> > function to handle daily returns aggregated to monthly and year? >
>> > >
>> It would be useful to have a table displaying Monthly Returns and
>> Total
>> > Return (rows), by year's (columns), for daily returns. > >
>> > Amarjit >
>> If an alternative package be acceptable as well, then function
>> 'returns'
>> in PMwR (which I maintain) might do what you want: returns(, period =
>> "month") in which series is a zoo (or xts) series. It is
>> described in the manual:
>> https://enricoschumann.net/R/packages/PMwR/manual/PMwR.html#holding-period-returns
>>
>> -- Enrico Schumann Lucerne, Switzerland https://enricoschumann.net
>>
>> [[alternative HTML version deleted]]
--
Enrico Schumann
Lucerne, Switzerland
https://enricoschumann.net
_______________________________________________
R-SIG-Finance@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should
go.
