Hello, if you have any need, please feel free to consult us, this is my wechat: wx91due
DATA4400 2024 Trimester 1: Assessment 1 (individual report 20 marks)
Go to Learning Forum> Week 5 TIME SERIES and download:
D44_A1_20240412_DATA
Open this file. The first four columns (A, B, C, D) are: t (month number); date; DonSales; WOTSales. There are 133 rows corresponding to labels plus 132 months (11 years) from January 2013 up to December 2023. The unit of measurement of sales is Lakh Erehwon $ per month (LE$/month).
Open a new Excel file and Copy & Paste123 the first four columns of D44_A1_20240412_DATA into your new file. Call this new file D44_A1_yourinitials. Your file will be different from anyone else’s.
Make two copies of your file. Call the first copy TRAIN and delete the rows corresponding to 2023. Call the second copy TEST and remove the rows corresponding to years 2013-2022, so that TEST has 13 rows: labels, plus monthly values for 2023.
You should submit your answers as a Word file. Figure labels are placed underneath the figure and table cations are placed above the table. Round the RMSE and MA SE to an appropriate number of significant figures. Parts a) and b) are for TheDon only. Parts c) d) e) refer to both stores. Part f) is for WorkOut only.
Copy and Paste, using source format, your TEST Excel sheet at the end of the Word file.
a) Plot TRAIN series plus HW forecasts for 2023 for TheDon in Tableau. Save the graph. Import the graph into your report and label it as Figure 1, with a detailed caption. [3 marks]
b) Plot TRAIN series plus Prophet forecasts for 2023 for TheDon in Exploratory. Save the graph. Import the graph into your report and label it as Figure 2, with a detailed caption. [3 marks]
c) Selecting from statistics given in Tableau and Exploratory, construct a table of RMSE within TRAIN series by method by store. [NOTE You need to make forecasts for WorkOut to obtain the RMSE and for Parts d) f), but you are not asked to provide graphs.]
Label this table as Table 1 and provide a detailed description. [4 marks]
d) Make calculations in Excel, or otherwise, construct a table of RMSE in TEST series by method by store. Label this as Table 2 and provide a detailed description. [4 marks]
e) Explain why the RMSE is expected to be larger in TEST series than it is in TRAIN series. Will the RMSE necessarily be larger in TEST series? AI generated answers not allowed. [3 marks]
f) Make calculations in Excel, or otherwise, find the MASE using Prophet in the test series for WorkOut. The mean absolute scaled error (MASE) is the ratio of the MAE using Prophet to the MAE using a naïve forecast. For this application define the naïve forecast as followsTake the naïve forecast for January 2023 as January 2022 sales, the naïve forecast for February 2023 as February 2022 sales, and so on until the naïve forecast for December 2023 is the December 2022 sales.
Copy and Paste, using source format, your TEST Excel sheet here. [3 marks]