Date Sales Jan-95 1664.81 Feb-95 2397.53 Mar-95 2840.71 Apr-95 3547.29 May-95 3752.96 Jun-95 3714.74 Jul-95 4349.61 Aug-
Posted: Fri Jul 01, 2022 5:51 am
Date
Sales
Jan-95
1664.81
Feb-95
2397.53
Mar-95
2840.71
Apr-95
3547.29
May-95
3752.96
Jun-95
3714.74
Jul-95
4349.61
Aug-95
3566.34
Sep-95
5021.82
Oct-95
6423.48
Nov-95
7600.6
Dec-95
19756.21
Jan-96
2499.81
Feb-96
5198.24
Mar-96
7225.14
Apr-96
4806.03
May-96
5900.88
Jun-96
4951.34
Jul-96
6179.12
Aug-96
4752.15
Sep-96
5496.43
Oct-96
5835.1
Nov-96
12600.08
Dec-96
28541.72
Jan-97
4717.02
Feb-97
5702.63
Mar-97
9957.58
Apr-97
5304.78
May-97
6492.43
Jun-97
6630.8
Jul-97
7349.62
Aug-97
8176.62
Sep-97
8573.17
Oct-97
9690.5
Nov-97
15151.84
Dec-97
34061.01
Jan-98
5921.1
Feb-98
5814.58
Mar-98
12421.25
Apr-98
6369.77
May-98
7609.12
Jun-98
7224.75
Jul-98
8121.22
Aug-98
7979.25
Sep-98
8093.06
Oct-98
8476.7
Nov-98
17914.66
Dec-98
30114.41
Jan-99
4826.64
Feb-99
6470.23
Mar-99
9638.77
Apr-99
8821.17
May-99
8722.37
Jun-99
10209.48
Jul-99
11276.55
Aug-99
12552.22
Sep-99
11637.39
Oct-99
13606.89
Nov-99
21822.11
Dec-99
45060.69
Jan-00
7615.03
Feb-00
9849.69
Mar-00
14558.4
Apr-00
11587.33
May-00
9332.56
Jun-00
13082.09
Jul-00
16732.78
Aug-00
19888.61
Sep-00
23933.38
Oct-00
25391.35
Nov-00
36024.8
Dec-00
80721.71
1-Jan
10243.24
1-Feb
11266.88
1-Mar
21826.84
1-Apr
17357.33
1-May
15997.79
1-Jun
18601.53
1-Jul
26155.15
1-Aug
28586.52
1-Sep
30505.41
1-Oct
30821.33
1-Nov
46634.38
1-Dec
104660.67
Date
Sales
Jan-95
1664.81
Feb-95
2397.53
Mar-95
2840.71
Apr-95
3547.29
May-95
3752.96
Jun-95
3714.74
Jul-95
4349.61
Aug-95
3566.34
Sep-95
5021.82
Oct-95
6423.48
Nov-95
7600.6
Dec-95
19756.21
Jan-96
2499.81
Feb-96
5198.24
Mar-96
7225.14
Apr-96
4806.03
May-96
5900.88
Jun-96
4951.34
Jul-96
6179.12
Aug-96
4752.15
Sep-96
5496.43
Oct-96
5835.1
Nov-96
12600.08
Dec-96
28541.72
Jan-97
4717.02
Feb-97
5702.63
Mar-97
9957.58
Apr-97
5304.78
May-97
6492.43
Jun-97
6630.8
Jul-97
7349.62
Aug-97
8176.62
Sep-97
8573.17
Oct-97
9690.5
Nov-97
15151.84
Dec-97
34061.01
Jan-98
5921.1
Feb-98
5814.58
Mar-98
12421.25
Apr-98
6369.77
May-98
7609.12
Jun-98
7224.75
Jul-98
8121.22
Aug-98
7979.25
Sep-98
8093.06
Oct-98
8476.7
Nov-98
17914.66
Dec-98
30114.41
Jan-99
4826.64
Feb-99
6470.23
Mar-99
9638.77
Apr-99
8821.17
May-99
8722.37
Jun-99
10209.48
Jul-99
11276.55
Aug-99
12552.22
Sep-99
11637.39
Oct-99
13606.89
Nov-99
21822.11
Dec-99
45060.69
Jan-00
7615.03
Feb-00
9849.69
Mar-00
14558.4
Apr-00
11587.33
May-00
9332.56
Jun-00
13082.09
Jul-00
16732.78
Aug-00
19888.61
Sep-00
23933.38
Oct-00
25391.35
Nov-00
36024.8
Dec-00
80721.71
1-Jan
10243.24
1-Feb
11266.88
1-Mar
21826.84
1-Apr
17357.33
1-May
15997.79
1-Jun
18601.53
1-Jul
26155.15
1-Aug
28586.52
1-Sep
30505.41
1-Oct
30821.33
1-Nov
46634.38
1-Dec
104660.67
As a souvenir shop at a beach resort town in Queensland,Australia is trying to forecast sales for the next 12 months (year2002) based on the monthly sales data (in Australian dollars)between 1995 and 2001. Data are available in SouvenirSales.csv.
1) Create a time plot of monthly sales for the souvenir shop.Provide your screenshot of the plot here. (1 point)
2) Based on the plot, do you see a trend in the data? If so,what kind of trend do you expect? (1 point)
3) Based on the plot, do you see a seasonality in the data? Ifso, how many months are in one “season ”? (1 point) Partition thedata into training and validation sets, with the validation setcontaining the last 12 months of data (year 2001). Remember to fitonly the training data.
4) Perform a naïve forecast with seasonal mean values (Note:naïve with seasonal mean is different from naïve forecast) (1point). Remember to fit only the training data. What is the RMSEfor the forecast on the validation data (Note: Look at the RMSEunder test set, not training set)? (1 point)
5) Run a regression model with a linear trend and monthlyseasonality. Remember to fit only the training data. a. What is theestimated trend coefficient? (1 point) What does this mean? (1point) b. What is the RMSE for the forecast on the validation data(0.5 point)? Does this model perform better or worse than the naïveforecast in Question 4? (0.5 point)
6) Use a Holt-Winter’s exponential smoothing approach to make aprediction. Remember to fit only the training data.
a. What is the RMSE for the forecast on the validation data (0.5point)? Does this model perform better or worse than the naïveforecast in Question 4 (0.5 point)?
b. Plot the predicted values and actual values of the validationdata. Provide your screenshot of the plot here. (1 point)
(Please answer all!) in R
Sales
Jan-95
1664.81
Feb-95
2397.53
Mar-95
2840.71
Apr-95
3547.29
May-95
3752.96
Jun-95
3714.74
Jul-95
4349.61
Aug-95
3566.34
Sep-95
5021.82
Oct-95
6423.48
Nov-95
7600.6
Dec-95
19756.21
Jan-96
2499.81
Feb-96
5198.24
Mar-96
7225.14
Apr-96
4806.03
May-96
5900.88
Jun-96
4951.34
Jul-96
6179.12
Aug-96
4752.15
Sep-96
5496.43
Oct-96
5835.1
Nov-96
12600.08
Dec-96
28541.72
Jan-97
4717.02
Feb-97
5702.63
Mar-97
9957.58
Apr-97
5304.78
May-97
6492.43
Jun-97
6630.8
Jul-97
7349.62
Aug-97
8176.62
Sep-97
8573.17
Oct-97
9690.5
Nov-97
15151.84
Dec-97
34061.01
Jan-98
5921.1
Feb-98
5814.58
Mar-98
12421.25
Apr-98
6369.77
May-98
7609.12
Jun-98
7224.75
Jul-98
8121.22
Aug-98
7979.25
Sep-98
8093.06
Oct-98
8476.7
Nov-98
17914.66
Dec-98
30114.41
Jan-99
4826.64
Feb-99
6470.23
Mar-99
9638.77
Apr-99
8821.17
May-99
8722.37
Jun-99
10209.48
Jul-99
11276.55
Aug-99
12552.22
Sep-99
11637.39
Oct-99
13606.89
Nov-99
21822.11
Dec-99
45060.69
Jan-00
7615.03
Feb-00
9849.69
Mar-00
14558.4
Apr-00
11587.33
May-00
9332.56
Jun-00
13082.09
Jul-00
16732.78
Aug-00
19888.61
Sep-00
23933.38
Oct-00
25391.35
Nov-00
36024.8
Dec-00
80721.71
1-Jan
10243.24
1-Feb
11266.88
1-Mar
21826.84
1-Apr
17357.33
1-May
15997.79
1-Jun
18601.53
1-Jul
26155.15
1-Aug
28586.52
1-Sep
30505.41
1-Oct
30821.33
1-Nov
46634.38
1-Dec
104660.67
Date
Sales
Jan-95
1664.81
Feb-95
2397.53
Mar-95
2840.71
Apr-95
3547.29
May-95
3752.96
Jun-95
3714.74
Jul-95
4349.61
Aug-95
3566.34
Sep-95
5021.82
Oct-95
6423.48
Nov-95
7600.6
Dec-95
19756.21
Jan-96
2499.81
Feb-96
5198.24
Mar-96
7225.14
Apr-96
4806.03
May-96
5900.88
Jun-96
4951.34
Jul-96
6179.12
Aug-96
4752.15
Sep-96
5496.43
Oct-96
5835.1
Nov-96
12600.08
Dec-96
28541.72
Jan-97
4717.02
Feb-97
5702.63
Mar-97
9957.58
Apr-97
5304.78
May-97
6492.43
Jun-97
6630.8
Jul-97
7349.62
Aug-97
8176.62
Sep-97
8573.17
Oct-97
9690.5
Nov-97
15151.84
Dec-97
34061.01
Jan-98
5921.1
Feb-98
5814.58
Mar-98
12421.25
Apr-98
6369.77
May-98
7609.12
Jun-98
7224.75
Jul-98
8121.22
Aug-98
7979.25
Sep-98
8093.06
Oct-98
8476.7
Nov-98
17914.66
Dec-98
30114.41
Jan-99
4826.64
Feb-99
6470.23
Mar-99
9638.77
Apr-99
8821.17
May-99
8722.37
Jun-99
10209.48
Jul-99
11276.55
Aug-99
12552.22
Sep-99
11637.39
Oct-99
13606.89
Nov-99
21822.11
Dec-99
45060.69
Jan-00
7615.03
Feb-00
9849.69
Mar-00
14558.4
Apr-00
11587.33
May-00
9332.56
Jun-00
13082.09
Jul-00
16732.78
Aug-00
19888.61
Sep-00
23933.38
Oct-00
25391.35
Nov-00
36024.8
Dec-00
80721.71
1-Jan
10243.24
1-Feb
11266.88
1-Mar
21826.84
1-Apr
17357.33
1-May
15997.79
1-Jun
18601.53
1-Jul
26155.15
1-Aug
28586.52
1-Sep
30505.41
1-Oct
30821.33
1-Nov
46634.38
1-Dec
104660.67
As a souvenir shop at a beach resort town in Queensland,Australia is trying to forecast sales for the next 12 months (year2002) based on the monthly sales data (in Australian dollars)between 1995 and 2001. Data are available in SouvenirSales.csv.
1) Create a time plot of monthly sales for the souvenir shop.Provide your screenshot of the plot here. (1 point)
2) Based on the plot, do you see a trend in the data? If so,what kind of trend do you expect? (1 point)
3) Based on the plot, do you see a seasonality in the data? Ifso, how many months are in one “season ”? (1 point) Partition thedata into training and validation sets, with the validation setcontaining the last 12 months of data (year 2001). Remember to fitonly the training data.
4) Perform a naïve forecast with seasonal mean values (Note:naïve with seasonal mean is different from naïve forecast) (1point). Remember to fit only the training data. What is the RMSEfor the forecast on the validation data (Note: Look at the RMSEunder test set, not training set)? (1 point)
5) Run a regression model with a linear trend and monthlyseasonality. Remember to fit only the training data. a. What is theestimated trend coefficient? (1 point) What does this mean? (1point) b. What is the RMSE for the forecast on the validation data(0.5 point)? Does this model perform better or worse than the naïveforecast in Question 4? (0.5 point)
6) Use a Holt-Winter’s exponential smoothing approach to make aprediction. Remember to fit only the training data.
a. What is the RMSE for the forecast on the validation data (0.5point)? Does this model perform better or worse than the naïveforecast in Question 4 (0.5 point)?
b. Plot the predicted values and actual values of the validationdata. Provide your screenshot of the plot here. (1 point)
(Please answer all!) in R