TR() function to calculate the true high, true low, and true range. Refactored ATR() to use the TR() function. Thanks to @openbmsjsc and Steve Bronder for the reports, and Ethan B. Smith for the PR. (#18, #114, #124)Fix stockSymbols() for ticker “NA”. read.table() converts the string “NA” to a missing value (NA) because na.strings = "NA" by default. This causes an issue because there’s actually a company with “NA” for the ticker. (#128)
CTI() did not pad its result with leading NA when the input was not coerced to an xts object. This was different from other TTR functions (e.g. SMA(), RSI(), ROC()). (#127)
Removed the VMA() function, which was never correct because the results made no sense.
Check that the wma() C function has enough non-NA values and throw an error if it doesn’t. This could cause the WMA() function to crash the user’s R session. (#126)
runMean(..., cumulative = TRUE) didn’t account for leading NA in the denominator. (#122)
runSD(x, cumulative = TRUE) returned all NA when x had any leading NA. Thanks to Ethan B. Smith for the report. (#121)
The TRIX() signal line did not use nSig unless maType was provided. Thanks to @SatoshiReport for the… report. (#120)
ALMA() calculation speed. Thanks to Ethan B. Smith for the report and suggested fix. (#117)runPercentRank() would segfault if x had fewer non-NA observations than the value for n. Thanks to Ian Fellows for the report. (#112)
run*(x, n = 1, cumulative = TRUE) functions would return NA for the first value. Thanks to Ethan B. Smith for the report and PR! (#111, #108, #88)
Fix NA check off-by-one error in aroon() that caused it to fail if there were exactly enough non-NA values. (#102)
ratio > 0 before calculating n in zlema() C code. The prior code could result in division by 0, which was flagged by clang-UBSAN. Thanks to Prof Brian Ripley for the report. (#100)Fix leading NA accounting in wma() C code. The prior code caused invalid reads under valgrind. Thanks to Prof Brian Ripley for the report. (#99)
Check for ratio > 0 before calculating n n ema() C code. The prior code could result in division by 0, which was flagged by UBSAN. Thanks to Prof Brian Ripley for the report. (#100)
Make ALMA() output length equal input length when the input can not be converted to xts. This was caused by the difference between rollapply.default() and rollapply.xts(). Thanks to GitHub user @marksimmonds for the report. (#29)
Fix stoch() in very rare cases where fastK = Inf. I could only reproduce this if the Close is > High and High and Low are equal, but that is a data error. I fixed anyway because there may be other cases I don’t anticipate. Thanks to GitHub user @cjuncosa for the report. (#52)
Fix MFI() when money flow is always zero or positive. The denominator of the money ratio will be zero if there is no negative money flow for n consecutive observations (e.g. during a strong up-trend), which causes the money flow index to be Inf. Set the money flow index to 100 in this case.
And the money ratio will be NaN if there’s no money flow for n consecutive observations (e.g. if there are no trades), which causes the money flow index to be NaN. Set the money flow index to 50 in this case.
Thanks to GitHub user @jgehw for the report, reproducible example, and suggested patch. (#81)
Updated stockSymbols() to use the NASDAQ FTP site instead of downloading the CSV from the NASDAQ stock screener page. Some columns are no longer populated because they are not provided in the FTP file: LastSale ,MarketCap, IPOyear, Sector, Industry These columns will be removed in a future version. (#98, #5, #97)
runPercentRank(x, n, cumulative = TRUE) now sets observations in the initialization period to NA. This is consistent with the other running/rolling functions in TTR. If you want the previous behavior, you should use runPercentRank(x, n = 1, cumulative = TRUE). Thanks to GitHub user @httassadar for the report. (#73)
runMAD() returned incorrect values when cumulative = TRUE and the input contained leading NA. Thanks to GitHub user @stellathecat for the report. This also affected runMedian(). (#93)
ZLEMA() would crash when ratio = 0.0 and n was not specified. Thanks to GitHub user @yogat3ch for the report! (#95)
WMA() did not return an xts object when passed an xts object for x that had leading NA, with the default wts = 1:n. Thanks to Cory Fletcher for reporting this issue via email. (#96)
stoch() was wrong when bounded = FALSE. Thanks to GitHub user @rfinfun for the report and patch. (#74)
HMA() threw an error when n was an odd number. This was because the first call to WMA() used n = n / 2 which caused n to not be an integer. Thanks to GitHub user @dragie for the report. (#76)
Update DVI() to use runPercentRank(). Thanks to Ivan Popivanov for the patch.
getYahooData() now returns an xts object with Date index (not POSIXct).
Column names for moving average function outputs are no longer based on the input column names.
Add HMA() and ALMA() functions/docs. Thanks to Ivan Popivanov.
Add Ultimate Oscillator function/docs/tests. Thanks to Ivan Popivanov.
run*() functions now error if there are not enough non-NA values.
Change all instances of lag() to lag.xts() in case x is a matrix. Thanks to Ivan Popivanov for the report.
Correct output column names in ATR() docs.
CLV() now sets NaN and Inf values to 0, instead of only NaN values.
Fix OBV() so OBV[t] = OBV[t-1] when Close[t] == Close[t-1].
Fix dead links in documentation.
CCI() now returns an object with column names (“cci”).
All moving average functions now attempt to set column names.
Added clarification on the displaced nature of DPO().
SAR() now sets the initial gap based on the standard deviation of the high-low range instead of hard-coding it at 0.01.
Added rollSFM() function that calculates alpha, beta, and R-squared for a single-factor model. Thanks to James Toll for the prototype.
Added runPercentRank() function. Thanks to Charlie Friedemann.
Moved slowest portion of aroon() to C.
DonchianChannel() gains an include.lag = FALSE argument, which includes the current period’s data in the calculation. Setting it to TRUE replicates the original calculation. Thanks to Garrett See and John Bollinger.
The Stochastic Oscillator and Williams’ %R now return 0.5 (instead of NaN) when a securities’ price doesn’t change over a sufficient period.
All moving average functions gain ....
Users can now change alpha in Yang Zhang volatility calculation.
Fixed MACD() when maType is a list. Now mavg.slow = maType[[2]] and mavg.fast = maType[[1]], as users expected based on the order of the nFast and nSlow arguments. Thanks to Phani Nukala and Jonathan Roy.
Fixed bug in lags(). Thanks to Michael Weylandt.
Corrected error in Yang Zhang volatility calculation. Thanks to several people for identifying this error.
Correction to SAR() extreme point calculations. Thanks to Vamsi Galigutta.
adjRatios() now ensures all inputs are univariate. Thanks to Garrett See.
EMA() and EVWMA() now ensure n is less than the number of non-NA values. Thanks to Roger Bos.
Fix to BBands() docs. Thanks to Evelyn Mitchell.
Fixed stockSymbols() for nasdaq.com changes (again), and attempted to make stockSymbols() more robust to nasdaq.com changes.
Corrected final calculation in Yang-Zhang volatility. Thanks to Shal Patel.
Corrected k in Yang-Zhang volatility. Thanks to Ian Rayner.
Corrected s2o and s2c in Yang-Zhang volatility. Thanks to Ian Rayner.
Corrected KST() when input is xts (result is now * 100). Thanks to Yuanwei.
Added variable moving average function, VMA().
Added Brian Peterson’s price bands function, PBands().
Added David Varadi’s DVI() indicator.
Added wilder and ratio arguments to DEMA. Thanks to Matthew Fornari for the suggestion.
Changed wilderSum() to seed initial value with raw sum. This matches Wilder’s original calculations. Thanks to Mahesh Bp for the report.
The BBands() standard deviation calculation now uses the population instead of sample statistic. This is consistent with Bollinger Band literature. Thanks to Jeff Ryan for the patch.
Fixed stockSymbols() for nasdaq.com changes.
Fixed ZLEMA() default ratio by changing it from 2/(n-1) to 2/(n+1). This makes it consistent with EMA(). Thanks to Dirk Eddelbuettel.
Corrected close-to-close volatility. Thanks to James Toll for the report.
adjRatios() failed (spectacularly) if there were missing close prices. Thanks to Garrett See for the report.
Added VWAP() and VWMA(). Thanks to Brian Peterson.
Added v-factor generalization to DEMA(). Thanks to John Gavin.
Updated volatility() to handle univariate case of calc = "close". Thanks to Cedrick Johnson.
Moved EMA(), SAR(), and wilderSum ()from .Fortran to .Call ()and used xts:::naCheck() instead of TTR’s NA check mechanism.
RSI ()up/down momentum is now faster with xts. Thanks to Jeff Ryan.
If ratio is specified in EMA ()but n is missing, the traditional value of n is approximated and returned as the first non-NA value.
Fix to stoch() when maType is a list and n is not set in the list’s 3rd element. Thanks to Wind Me.
Fixed fastK in stoch() when smooth != 1.
Fixed segfault in EMA ()when n < NROW(x). Thanks to Douglas Hobbs.
test.EMA.wilder() failed under R-devel. Thanks to Prof Brian Ripley.
CMO(), DPO(), DonchianChannel(), RSI(), and TDI () to explicitly use xts internally.Fixed bug in WMA(), EVWMA(), ZLEMA(), and GMMA(); results were not being reclass()ed back to their original class.
cbind () call in the following functions:
ADX()aroon()ATR()BBands()DonchianChannel()EMV()KST()MACD()stoch()SMI()TDI()TRIX()Fixed bug in VHF(); missing abs() calculation in the denominator. Thanks to Jürgen Wurzer for the report!
getYahooData() now returns an xts object.
Added column names to output for ADX(), EMV(), and CLV () (for xts).
momentum() in CMO() no longer sets na = 100.
Replaced na argument in momentum() and ROC() with na.pad.
maType argument default values from function formals to function body for the following functions:
ADX()ATR()CCI()DPO()EMV()KST()MACD()RSI()TRIX()BBands()chaikinVolatility()stoch()SMI()adjRatios() creates split and/or dividend adjustment ratio series via C code.
GMMA() calculates the Guppy Multiple Moving Average.
volatility() now has Yang Zhang, and Garman-Klass (Yang Zhang) calculations.
runSum(), runMin(), runMax()runMean(), runMedian()runCov(), runCor(), runVar(), runSD(), runMAD()Added internal smoothing to FastK in stoch() via smooth argument. Thanks to Stanley Neo.
getYahooData() now uses adjRatios(), which yields significant speed improvements for larger data sets.
All functions now use xts internally, adding support for all major time series classes. If try.xts() fails on the input object(s), they will be converted to a matrix and a matrix object will be returned.
Added bounded arg to stoch() and SMI(), which includes the current period in the calculation.
Added the zig zag indicator: ZigZag().
volatility(), with the following calculations:
Added Money Flow Index: MFI().
Added Donchian channel: DonchianChannel().
Added multiple argument to TDI(), allowing more user control.
Added naCheck() and implemented it in the moving average functions.
Fixed bug when maType was a list and n was not specified in maType. This affected: stoch(), SMI(), RSI(), KST(), MACD(), TRIX().
Corrected NaN replacement in CLV().
Corrected williamsAD(): the result is 0 if C(t) = C(t-1).
Corrected runMedian() and runMAD(). The argument controlling which type of median to calculate for even-numbered samples wasn’t being passed to the Fortran routine.
aroon() calculation starts at period n+1, instead of n.
Added NA to first element of closeLag of ATR().
Corrected BBands() and CCI() for rowMeans() use on xts objects.
Made changes to Rd files to pass R CMD check on R-devel (2.9.0).
Changed default type of ROC() to continuous.
Changed BBands() %B output value from pct.b to pctB.
Changed WPR() output value from pct.R to pctR.
Changed WPR() MA output value from ma.emv to emvMA.
Changed aroon() output values from aroon.xx to aroonXx.
chaikinMF() to CMF()stochastic() to stoch()bollingerBands() to BBands()Set na = NA for momentum() and ROC() functions in files KST.R, RSI.R, and TDI.R, and changed ROC() to use type = "discrete" in chaikinVolatility.R.
ZLEMA() function:
ratio = NULL argument.ratio argument.Changed the BBands() function’s sd argument from a list that allows other dispersion functions to simply indicate the number of standard deviations to use.
maType arguments and updated documentation for:
RSIm()ADX()ATR()CCI()DPO()EMV()RSI()BBands()chaikinVolatility()stoch()SMI()TRIX()MACD()KST()Added Stochastic Momentum Index SMI() and williamsAD() functions and documentation.
SMA()EMA()WMA()EVWMA()ZLEMA()PSAR()Added NA checking/handling for many functions.
Added ratio = NULL argument to EMA().
roll*() to their respective Fortran implementations and removed the rollFun() function. Added Fortran based functions are:
runSum()wilderSum()runMin()runMax()runMean()runCov()runCor()runVar()runSD()runMedian()runMAD()Changed CCI() to use runMAD() internally.
Removed oscillator() function and transferred functionality to MACD() function.
Removed chaikinOscillator(), since it can be created via MACD(chaikinAD(...)).
match.arg(type) in ROC() changed to simple subsetting of type.
Changed trailing zeros to trailing NAs in DPO().
Fixed WMA() bug that allowed x and wts vectors to have different length if either series had leading NAs (similar to EVWMA() function).
Fixed runCov() bug that allowed x and y vectors to have different length if either series had leading NAs (similar to EVWMA() function).
Corrected EVWMA() to start at period n instead of n-1.
Removed message function from CCI.R, VHF.R, WPR.R, aroon.R bollingerBands.R, and stochastics.R.
Changed order of oscillator() arguments from ma.slow, ma.fast, ma.sig to the traditional ma.fast, ma.slow, ma.sig. Thanks to Jeff Ryan.
The arguments to the chaikinOscillator() function were changed as above.
Changed EVWMA() so period n contains the value for periods (i-n+1):n and so periods 1:(n-2) will be NA.
Changed EMA() so periods 1:n will be NA.
Changed bbands() to bollingerBands().
Changed DX() to ADX().
Changed stoch() to stochastic().