Exploring portfolio-based conjectures about financial instruments

Description

The backtest package provides facilities for exploring portfolio-based conjectures about financial instruments (stocks, bonds, swaps, options, et cetera).

Details

Index:

backtest                Creating an Object of Class Backtest
backtest-class          Class "backtest"
starmine                StarMine Rankings, 1995

Further information is available in the following vignettes:

Author(s)

Jeff Enos <jeff@kanecap.com> and David Kane <dave@kanecap.com>, with contributions from Kyle Campbell <kyle.w.campbell@williams.edu>, Daniel Gerlanc <daniel@gerlanc.com>, Aaron Schwartz <Aaron.J.Schwartz@williams.edu>, and Daniel Suo <danielsuo@gmail.com>

Maintainer: Jeff Enos <jeff@kanecap.com>

Creating an Object of Class Backtest

Description

Conducts a backtest and returns the results as an object of class backtest.

Usage

backtest(x,
         in.var,
         ret.var,
         universe,
         by.var    = NULL,
         date.var  = NULL,
         id.var    = NULL,
         buckets   = 5,
         natural   = FALSE,
         do.spread = TRUE,
         by.period = TRUE,
         overlaps  = 1)

Arguments

Details

Data frames for backtest must, at a minimum, contain a column of class numeric to be referenced by the in.var and ret.var arguments.

The in.var is the primary variable by which the backtest categorises observations. It must reference a numeric column in x. Using the values in x, backtest breaks the values into equal sized quantiles, or buckets.

The by.var is the secondary variable by which the backtest categorises observations. When specifying both in.var and by.var, backtest organises the observations into a n by j matrix where n is the number of quantiles or categories created for the by.var and j is the number of quantiles created for the in.var. By default, backtest creates 5 quantiles.

If natural is TRUE, the data and arguments must meet certain requirements. First, the frequency of the observations and ret.var must be the same. Second, an id.var and date.var are required. Third, a by.var is not allowed. Note that the code does not verify that the backtest is truly natural; backtest accepts the value passed by the user as valid.

Value

Returns an object of class backtest.

The functions show and summary are used to obtain and print a short description and longer summary of the results of the backtest. The accessor functions counts, totalCounts, marginals, means, naCounts, and turnover extract different parts of the value returned by backtest.

Author(s)

Kyle Campbell kyle.w.campbell@williams.edu and Jeff Enos jeff@kanecap.com

See Also

backtest-class

Examples

data(starmine)

## Backtest with 1 'in.var' and 1 'ret.var'

bt <- backtest(starmine, in.var = "smi", ret.var = "ret.0.1.m", by.period = FALSE)
summary(bt)

## Backtest with 2 'in.var' values, 1 'ret.var', and a 'by.var'

bt <- backtest(starmine, in.var = c("smi", "cap.usd"),
               ret.var = "ret.0.1.m", by.var = "sector", by.period = FALSE)
summary(bt)

## Backtest with 1 'in.var', 1 'by.var', and 1 'ret.var'.  Number of
## buckets changed from default of 5 to 4.  Change in number of buckets
## only affects the 'in.var' because the 'by.var' column in 'starmine'
## contains character data. For each value in this column there is a
## unique category.

bt <- backtest(starmine, in.var = "smi", by.var = "sector",
               ret.var = "ret.0.1.m", buckets = 4, by.period = FALSE)
summary(bt)

## Backtest with 1 'in.var', multiple 'ret.var', and a
## universe restriction

bt <- backtest(starmine, in.var = "smi",
               ret.var = c("ret.0.1.m", "ret.0.6.m"),
               universe = sector == "HiTec", by.period = FALSE)
summary(bt)

## Running a natural backtest with 2 'in.vars', 1 'ret.var'
## 10 buckets

bt <- backtest(starmine, in.var = c("smi","cap.usd"),
               ret.var = "ret.0.1.m", date.var = "date",
               id.var = "id", buckets = 10,
               natural = TRUE, by.period = FALSE)
summary(bt)

## The same backtest, but calculating quantiles within periods.

bt <- backtest(starmine, in.var = c("smi","cap.usd"),
               ret.var = "ret.0.1.m", date.var = "date",
               id.var = "id", buckets = 10,
               natural = TRUE, by.period = TRUE)
summary(bt)

plot(bt, type = "turnover")
plot(bt, type = "return")
plot(bt, type = "cumreturn")
plot(bt, type = "cumreturn.split")

Class "backtest"

Description

Contains results from the backtest function.

Details

The primary method for accessing the backtest results is through the summary method. summary provides different displays depending on the type of backtest object. These displays are shown in the examples section. Accessor methods such as means, counts, marginals, naCounts, turnover, and ci may be used to extract other types of information from the object.

A backtest object with a natural value of TRUE may be graphed by calling the plot method. The default plot method graphs return. The other plots, turnover and cumulative return, must be explicitly specified as plot(object, type = "turnover") or plot(object, type = "cumreturn").

The backtest object does not store the data frame used to create the backtest. It only stores the results and the names of the vectors used in calculating these results.

The results of a backtest are stored in a 5-dimensional array, results. The 1st dimension contains one value for every element of ret.var. The 2nd dimension contains one value for every element of in.var. The 3rd dimension contains one value for every element in 1:buckets[1], a vector from 1 through the number of by.var buckets. The 4th dimension contains one value for every element in 1:buckets[2], a vector from 1 through the number of in.var buckets. The 5th dimension contains 4 elements: means, counts, trim.means, and NAs.

Objects from the Class

Objects can be created by calls to the function backtest(data, in.var, ret.var, ...).

Slots

in.var:

Object of class "character" specifying the in.var values for this backtest.

ret.var:

Object of class "character" containing the ret.var values for this backtest.

by.var:

Object of class "character" containing the by.var, if specified, for this backtest.

date.var:

Object of class "character" containing the date.var, if specified, for this backtest.

buckets:

Object of class "numeric" containing the number(s) of buckets used create quantiles from the in.var and by.var values.

results:

A 5-dimensional "array" containing the results of the backtest.

ret.stats:

Object of class "array" containing return statistics for the backtest.

turnover:

Object of class "array" containing turnover statistics for the backtest.

natural:

Object of class "logical" expressing whether or not the intervals between observations, as specified by date.var, and returns, as specified by ret.var, match. If the interval between dates is one month, the interval between returns should also be one month.

do.spread:

Object of class "logical". If TRUE the summary method displays information about the spread between the extreme quantiles. If FALSE this information is suppressed. Defaults to TRUE.

by.period:

Object of class "logical". If TRUE the quantiles are recalculated within each date period. If FALSE the quantiles are calculated all at once. Defaults to TRUE.

overlaps:

An object of class "numeric" which specifies the number of prior periods to include in the current period's portfolio weights calculation. If overlaps is the default of 1, backtest behaves as usual and only uses a periods own data to determine its portfolio. If overlaps is set to n > 1, a period's portfolio comprises the weighted mean of portfolio weights from the previous n periods, with period n having a weight of 1/n.

Methods

show

signature(object = "backtest"): Prints the variables used in this backtest.

summary

signature(object = "backtest"): Prints the results of the backtest.

summaryStats

signature(object = "backtest"): Returns a data frame with spreads for each date.var value and each in.var.

means

signature(object = "backtest"): Returns a list of matrices, with one matrix for each in.var, where the value of each cell is the mean of the returns for that in.var and by.var combination.

counts

signature(object = "backtest"): Returns a list of matrices, with one matrix for each in.var, where the value of each cell is the number of observations for that in.var and by.var combination.

totalCounts

signature(object = "backtest"): Returns a data frame in the same format as the speads data frame returned by summaryStats: contains the sum of counts for all buckets (or high and low buckets if argument low.high.only is set to TRUE) of non-NA in.var values that went into the spread calculations.

marginals

signature(object = "backtest"): Returns a list of matrices, one matrix for each in.var, where the value of each cell is the number of observations for that in.var and by.var combination. Different from counts because the marginal sums have been appended to the matrices.

naCounts

signature(object = "backtest"): Returns a list of matrices, with one matrix for each in.var, where the value of each cell is the number of NA observations for that in.var and by.var combination.

ci

signature(object = "backtest"): Returns a matrix of confidence intervals for spreads.

turnover

signature(object = "backtest"): Returns a data.frame of the turnovers if the backtest is natural.

plot

signature(x = "backtest", y = "missing"): Plots returns, cumulative returns, or turnover, when passed a type argument of return, cumreturn, or turnover, respectively.

Author(s)

Kyle Campbell kyle.w.campbell@williams.edu

See Also

backtest

Examples

data(starmine)
bt <- backtest(starmine, in.var = "smi", ret.var = "ret.0.1.m", by.period = FALSE)

## Summary for a pooled backtest

summary(bt)

## A natural backtest

bt <- backtest(starmine, in.var = "smi", ret.var = "ret.0.1.m",
               date.var = "date", id.var = "id", natural = TRUE, by.period = FALSE)

## Summary for a natural backtest

summary(bt)

## Other access methods

means(bt)
counts(bt)
marginals(bt)
naCounts(bt)

## Plotting methods

plot(bt, type = "turnover")
plot(bt, type = "return")
plot(bt, type = "cumreturn")

StarMine Rankings, 1995

Description

StarMine rankings of some stocks in 1995, with corresponding returns and other data.

Usage

data(starmine)

Format

A data frame containing 53328 observations on the following 23 variables.

date

Date on which the observation was recorded. The dates have a monthly frequency. Dates range from 1995-01-31 to 1995-11-30.

id

Unique identifier for each stock.

symbol

Company symbol.

name

Full company name.

country

Country of the exchange on which the company is listed. This factor has levels AUS, CHE, DEU, DNK, ESP, FIN, FRA, GBR, HKG, ITA, JPN, NLD, NOR, NZL, SWE and USA

sector

Sector specification. This factor has levels Durbl, Enrgy, HiTec, Hlth, Manuf, Money, NoDur, Other, Shops, Telcm and Utils

sec

An alternative sector specification. This factor has levels CND, CNS, COM, ENE, FIN, HTH, IND, MAT, TEC and UTL.

ind

Industry specification. This factor has levels AERDF, AIRLN, AUTOP, AUTOS, BANKS, BEVGS, BIOTC, BUILD, CHEMS, CNENG, CNFIN, CNMAT, COMEQ, COMPT, COMSS, CONGL, CPMKT, DICNS, DISTR, DVFIN, DVTEL, ELEQI, ELEQT, ELUTL, ENEQS, FDPRD, FDRET, GSUTL, HEPSV, HEQSP, HETEC, HOTEL, HSDUR, HSPRD, INSUR, INTSS, IPPET, ITCAT, ITCON, LEISR, LFSCI, LOGIS, MACHN, MEDIA, METAL, MGFIN, MLRET, MLUTL, OFFIC, OILGS, PACKG, PAPER, PHARM, PRPRD, REALE, REDEV, REITS, RRAIL, SEMIP, SEMIS, SHIPS, SMOKE, SOFTW, SPRET, TEXAP, TRADE, TRINF, WIREL and WTUTL

size

cap.usd normalized to N(0,1).

smi

StarMine Indicator (smi) score

liq

Liquidity of the company.

ret.0.1.m

One-month forward return of the company.

ret.0.6.m

Six-month forward return of the company.

ret.1.0.m

One-month prior return of the company.

ret.6.0.m

Six-month prior return of the company.

ret.12.0.m

Twelve-month prior return of the company.

mn.dollar.volume.20.d

Mean dollar volume of the past 20 days.

md.dollar.volume.120.d

Median dollar volume of the past 120 days.

cap.usd

Market capitalisation of the company in USD.

cap

Market capitalisation of the company in local currency.

sales

Annual gross sales of the company.

net.income

Annual net income of the company.

common.equity

Annual common equity of the company.

Details

starmine contains selected attributes such as sector, market capitalisation, country, and various measures of return for a universe of approximately 6,000 stocks. The data is on a monthly frequency from January 31, 1995 to November 30, 1995.

Note

We would like to thank StarMine Corporation for allowing us to include this data in the backtest package.

Source

StarMine Corporation. For more information, see http://www.starmine.com.

Examples

data(starmine)
head(starmine)