HoltWinters - output of original and fitted values in a csv file (R)

Question

I have daily temperature data for several years. I use the HoltWinters method to conduct exponential smoothing for the existing dataset. No forecast.

I need to find a convenient way to extract the original data and fitted data (xhat) for some manual analysis. .csv file would be ideal.

Since I am new to R, I am not sure how to extract the data in the needed format.

1. Sample input data

      02/01/1998  98
      02/02/1998  100
      02/03/1998  96

after displaying:

    DAY   1998 1999 2000 2001 2002 2003

1   1-Jul   98  86  91  84  89  84  
2   2-Jul   97  90  88  82  91  87  

2. Desired Output (the last number is a fitted result)

   02/01/1998 98 97.6 02/02/1998 100 100.4 02/03/1998 96 95.7

3. My code

ts1 <- read.table("ts.txt", header=TRUE)
ts_v <- as.vector(unlist(ts1[,2:21]))
ts_ts <- ts(data = ts_v, frequency=123, start=1998)
hw_ts_add <- HoltWinters(ts_ts, seasonal = "additive")

print(hw_ts_add$fitted) 


          xhat     level        trend        season
1997.000  87.17619  82.87739 -0.004362918   4.303159495
1997.008  90.32925  82.09550 -0.004362918   8.238118845
1997.016  92.96089  81.87348 -0.004362918  11.091777381
1997.024  90.93360  81.89497 -0.004362918   9.042996893
1997.033  83.99752  81.93450 -0.004362918   2.067387137

How Do I extract the date, xhat and the initial data as:

    date  xhat  orig

Answer 1

You can extract xhat by using

print(hw_ts_add$fitted[,1])

To extract the first column of that matrix. Sadly, you cannot use hw_ts_add$fitted$xhat