Errors when using getSymbols() quantmod to predict stock price - r

Question

I was trying to use quantmod to download some history data of stock price, here's my code:
Nasdaq100_Symbols <- c('GE','PG','MSFT','AAPL','PFE','AMD','DELL','GRPN','FB','CSCO','INTC',
'EZJ.L','BP','HSBC','MKS')
getSymbols(Nasdaq100_Symbols)
Warning messages:
1: DELL contains missing values. Some functions will not work if objects contain missing values in the middle of the series. Consider using na.omit(), na.approx(), na.fill(), etc to remove or replace them.
How can I remove these NA values since I'm trying to merge them together and make it as a time series data type,
nasdaq100 <- data.frame(as.xts(merge(GE,PG,MSFT,AAPL,PFE,AMD,DELL,GRPN,FB,CSCO,INTC,
EZJ.L,BP,HSBC,MKS)))
head(nasdaq100[,1:12],2)
GE.Open GE.High GE.Low GE.Close GE.Volume GE.Adjusted PG.Open PG.High PG.Low
2007-01-02 NA NA NA NA NA NA NA NA NA
2007-01-03 37.41 38.15 37.38 37.97 43222800 24.48669 63.72 64.66 63.7
PG.Close PG.Volume PG.Adjusted
2007-01-02 NA NA NA
2007-01-03 64.54 9717900 44.56958
class(nasdaq100)
[1] "data.frame"
# set outcome variable
outcomeSymbol <- 'FISV.Volume'
# shift outcome value to be on same line as predictors
library(xts)
nasdaq100 <- xts(nasdaq100,order.by=as.Date(rownames(nasdaq100)))
nasdaq100 <- as.data.frame(merge(nasdaq100,lm1=lag(nasdaq100[,outcomeSymbol],-1)))
Error in `[.xts`(nasdaq100, , outcomeSymbol) : subscript out of bounds
I'm stuck here, I found a tutorial on Youtube(https://www.youtube.com/watch?v=lDgvaJFpybU&t=32s) but can't move forward because of these warning and errors, can someone tell me how to fix it?

Answer 1

If you are going to do part of an example code, make sure you adjust everything correctly. At the end you are filling the outcomeSymbol with a value from the stock FISV that you didn't download in the beginning of your script. And I must say the code in the script that you can find here could be written better. There are way too many switches between xts and data.frame that are not necessary. I'm not going to rewrite his whole code. But this code fixes your errors.
First, instead of polluting your work environment with a 100 stocks I put everything in one list object. Then merge all this together with Reduce and merge. The missing data that is in the DELL ticker, will nicely merge with everything else, but will be NA as there is no data. If you want to deal with this, either do not download the DELL data, or fill it with 0 with the na.fill function. This last option might not be a good solution if you are going to use this data for training a model. I also show you how to turn a xts object into a data.frame without having to use as.Date later on.
library(quantmod)
Nasdaq100_Symbols <- c('GE','PG','MSFT','AAPL','PFE','AMD','DELL')
# put all stocks in one list object
stocks <- lapply(Nasdaq100_Symbols, getSymbols, auto.assign = FALSE)
# following is not needed but if you want to use the list for other purposes
# it is a good practice to name all the different list objects.
# names(stocks) <- Nasdaq100_Symbols
# merge all stocks into 1 xts object
nasdaq100 <- Reduce(merge, stocks)
# fill NA's with 0
nasdaq100 <- na.fill(nasdaq100, 0)
outcomeSymbol <- "GE.Volume" # <-- used GE as that data is available in the downloaded data set
# merge outcome to data
nasdaq100 <- merge(nasdaq100, lm1 = lag(nasdaq100[, outcomeSymbol], -1))
# turn into data.frame
nasdaq100_df <- data.frame(date = index(nasdaq100), coredata(nasdaq100))

Answer 2

I am not exactly sure why you want to remove NA before merge.
I do it after merge and it works perfectly for me because xts objects are merged based on their data index. I only keep Adjusted Clase so my usual code looks like:
yahoo_symbols <- c(share1, share2, share3,...)
qts_env <- new.env()
getSymbols(yahoo_symbols,
env = qts_env,
from = start_date,
to = end_date,
periodicity = "daily"
)
shares_cl <- do.call(merge, eapply(qts_env, Ad))
shares_cl <- na.omit(shares_cl)
I hope that it helps.