The dataset store$\_$sales.csv is a subset of data from $2010$ to $2012$ from a US based retailer. It represents weekly sales. Each row is a different week of data for a given store and department for the included time frame.

Notes: The dataset includes periods from $2010$ to $2012,$ but it does not include every week. Do NOT extrapolate to fill in these missing pieces unless specifically instructed as it will change the answer. We will assume we are using a full dataset for most of the questions.

store $=$ store location id

dept $=$ store department id

date $=$ first day of given week $($string format you will need to parse it$)$

weekly$\_$sales $=$ sales for that week, dept, store

isholiday $=$ true if it is holiday week Question: Create a time$-$series dataset by computing the total weekly$\_$sales $($sum across all departments$)$ per week for store $1,$ do not exclude holidays. With this data set which we will refer to as "Store $1$ weekly sales", use an alpha of $0\mathrm{.}25$ and h $=100,$ and the ses$()$ function from the 'forecast' package to fit an exponential smoothing model on the data. Do not separate train$/$test sets or holdout any points, instead fit the model on the full period of the "Store $1$ weekly sales" dataset. Calculate and report the tracking signal based on the actual vs$.$ fitted values and report it rounded to $2$ decimal places. Provide code in R