Comparing two columns in xts object

Question

I want to pairwise compare two columns of an xts object. The xts object is a merged object from two individual xts vectors:

Results<-na.omit(merge(NCG,SMA_10))

Here is the structure of the resulting xts object:

An ‘xts’ object on 2016-04-28/2017-01-01 containing:
  Data: num [1:171, 1:2] 1.47 1.55 1.54 1.55 1.41 ...
 - attr(*, "dimnames")=List of 2
  ..$ : NULL
  ..$ : chr [1:2] "NCG" "SMA"
  Indexed by objects of class: [POSIXlt,POSIXt] TZ: 
  xts Attributes:  
List of 1
 $ na.action:Class 'omit'  atomic [1:19] 1 2 3 4 5 6 7 8 9 10 ...
  .. ..- attr(*, "index")= num [1:19] 1.46e+09 1.46e+09 1.46e+09 1.46e+09 1.46e+09 ...

The comparison is just toying around. Here it is:

for (i in 25:nrow(Results))
  {
   if(Results$NCG[i]>Results$SMA_10[i])
  {
    Results$Signal[i]="buy"
    }

    }

The error this code produces is the following:

Error in .xts(e, .index(e1), .indexCLASS = indexClass(e1), .indexFORMAT 

= indexFormat(e1),  :   index length must match number of observations 

Does anyone have an idea why this does not work? I suspect the removed values from the two initial xts objects....

Answer 1

You're trying to do an index operation, Results$Signal[i], on a column named Signal which you haven't yet defined.

Create the column Signal before you try to access and update an element in it. Here's one approach:

# Create the column with default values
Results$Signal <- NA

for (i in 25:nrow(Results))
{
    if(Results$NCG[i]>Results$SMA_10[i])
    {
        Results$Signal[i]="buy"
    }

}

(Note that your example would be much more efficiently solved using vectorisation and not an explicit for loop)