Stock Technical Indicators MFI error Series contains non-leading NAs

I am creating stock technical indicators for technical analysis. Most of the indicators work fine except MFI. The moment code reaches to MFI, I get below error

Series contains non-leading NAs

I have verified that there are no non leading NAs in stock series.Still I get this error. Below is a reproducible code. Appreciate any help.

library(ggplot2) 
library(quantmod) 
library(xts) 
library(timeSeries) 
library(rpart)
library(ROCR)


get_indicators <- function(stock, period){
  stock <- na.locf(stock)
  price_change <- Ad(lag(stock,-period)) - Ad(stock)
  response <- ifelse(price_change > 0, "Up", "Down")
  
  #Simple Moving Average
  sma15 <- SMA(Ad(stock),n=15)
  sma50 <- SMA(Ad(stock),n=50)

  #Exponential Moving Average
  ema15 <- EMA(Ad(stock),n=15)
  ema50 <- EMA(Ad(stock),n=50)
  
  #Moving Average Convergence Divergence
  macd <- MACD(Ad(stock), nFast=12, nSlow=26, nSig=9, maType= "EMA")  

  #Relative Strength Index
  rsi <- RSI(Ad(stock), n=14, maType= "EMA")
  
  #High, Low, and Adjusted Close xts object
  hlac <- as.xts(data.frame(x=Hi(stock), y=Lo(stock), z=Ad(stock)))
  
  #Stochastic Oscillator
  sto <- stoch(hlac, nFastK = 14) *100
  
  #Commodity Channel Index
  cci <-CCI(hlac, n = 20, c = 0.015)
  
  #Price Rate of Change
  proc <- ROC(Ad(stock), n=14) *100
  
  #Momentum
  mom <- momentum(Ad(stock), n=14)
  
  #Bollinger Bands
  bb <- BBands(Ad(stock),n=20,sd = 2)
  
  #Average True Range
  atr <- ATR(hlac, n=14)
  
  #On Balance Volume
  obv <- OBV(Ad(stock), Vo(stock))
  
  print(anyNA(stock))
  print(sum(is.na(stock)))
  
  
  #Money Flow Index
  mfi <- MFI(hlac,Vo(stock),n=14)
  
  #Feature Aggregation: Combine all derived indicators for model buuilding
  indicators <- data.frame(sma15, sma50, ema15, ema50, macd, rsi, 
                           sto, cci, proc, mom, bb$mavg, atr$atr, obv,  response)
  colnames(indicators) <- c("SMA.15","SMA.50","EMA.15","EMA.50","MACD","MACDSignal",
                            "RSI","StoFASTK","StoFASTD","StoSLOWD", "CCI.20","PROCR",
                            "MOM.14", "BB","ATR", "OBV","Response")
  
 
  return(indicators)
}
start.date <-'2015-01-01'
end.date <-'2020-01-01'
quantmod::getSymbols("SBIN.NS",src="yahoo",from=start.date,to=end.date,periodicity="daily")
stoci <- get_indicators(SBIN.NS,20)
stoci

Solution

The problem seems to be with your hlac object. Using adjusted closing price instead of actual closing price might be the reason.

The problem code works fine when using a quantmod::HLC() object instead.

``` r
library(quantmod)
#> Loading required package: xts
#> Loading required package: zoo
#> 
#> Attaching package: 'zoo'
#> The following objects are masked from 'package:base':
#> 
#>     as.Date, as.Date.numeric
#> Loading required package: TTR
#> Registered S3 method overwritten by 'quantmod':
#>   method            from
#>   as.zoo.data.frame zoo
#> Version 0.4-0 included new data defaults. See ?getSymbols.

from <-'2015-01-01'
to <-'2020-01-01'
stock <- "SBIN.NS"

getSymbols(stock, from = from, to = to, src = 'yahoo')
### Messages removed ###
#> [1] "SBIN.NS"

SBIN.NS <- na.locf(SBIN.NS)
sum(is.na(SBIN.NS))
#> [1] 0

#hlac <- as.xts(data.frame(x=Hi(SBIN.NS), y=Lo(SBIN.NS), z=Ad(SBIN.NS)))

hlc <- HLC(SBIN.NS)

mfi <- MFI(hlc,Vo(SBIN.NS),n=10)

head(mfi, n = 20)
#>                 mfi
#> 2015-01-01       NA
#> 2015-01-02       NA
#> 2015-01-05       NA
#> 2015-01-06       NA
#> 2015-01-07       NA
#> 2015-01-08       NA
#> 2015-01-09       NA
#> 2015-01-12       NA
#> 2015-01-13       NA
#> 2015-01-14       NA
#> 2015-01-15 53.58373
#> 2015-01-16 41.74790
#> 2015-01-19 41.48900
#> 2015-01-20 51.62824
#> 2015-01-21 64.43703
#> 2015-01-22 65.38013
#> 2015-01-23 74.19176
#> 2015-01-27 75.15463
#> 2015-01-28 76.77622
#> 2015-01-29 73.79739

^{Created on 2020-12-22 by the reprex package (v0.3.0)}