DSCI 2710 CASE ASSIGNMENT 2 A Decomposition Model for Sales 1. Introduction Cost
ID: 2929461 • Letter: D
Question
DSCI 2710 CASE ASSIGNMENT 2 A Decomposition Model for Sales 1. Introduction Cost-Mart is a national chain of department stores, offering groceries, clothing, consumer electronics, small appliances, and light furniture. Sales data (in US dollars) for 247 consecutive days, Jan 2 to Sep 5, 2015, were obtained. You are asked to fit a multiplicative time series decomposition model. Cost-Mart is particularly interested in the time trend and seasonal indices components. They would also like to use your model and provide sales forecasts for the next 7 days after the end of the provided data set, i.e., Sep 6 to Sep 12. 2. The data The daily sales data are provided in Excel file Case2_StoreSales.xls. The Sales column is of particular interest. The file is posted on our course Blackboard site. 3. Time Series Plot in Excel Select the 247 rows A2:B248, in columns A and B. Then select Insert>Insert Line Chart>2-D Line. The line chart will appear. Edit the chart title to something meaningful. 44 11 4) FS Fo F10 5 0Explanation / Answer
1)
The meaning of trend equation as the name suggests takes the data points and finds out the line that best fits the data to reveal a certain trend or pattern in the data given.
2)
Seasonal indices takes into account the season change and how a variable is affected in the particular season.
For example a certain item sells more in summer but will not be selling in the same proportion in other seasons.
This particular trend is captured or reflected in the seasonal index.
3)
The first forecast is
285024+263.1*t=Yt
here t=1 for september 6, 2015
=285024+263.1*1=285287.1 but the actual sales was 378,732.27 almost 100,000 more than predicted.
Hence we can say that the prediction can give us an estimate of sales but not the exact number.
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.