Thread: [SOLVED] Decisions and Forcasting

Hi everybody,

I am doing the quantitative methods project. It is mainly concerned with the forecasts and decisions. The only data that i have got were sales for 20 different time periods. I have done different forcasting methods and now i have chosen the "best one" . The one with smallest mean square error. In my case it was a linear trend model. And i got the following data for the forcasted sales (also for the next period). From the data below i need to make a discussion of - the forecasts and suggestions about the future in terms of advertising, marketing etc. As i have already mentioned i have also made other forecasting methods (Moving averages 2, MA3 and MA4 , together with linear trend, multiplicative model) I have made a graphs for each model so that i can more easily try to make a suggestions for the future!

Here is the graph for the best forcasted method (linear trend)



*****************************

THE LINEAR TREND EQUATION:


T = 33072.737 + 63.549 t

where T = trend value of the time series in period t



TIME PERIOD TIME SERIES VALUE FORECAST FORECAST ERROR
=========== ================= ======== ==============

1 32,340 33,136.29 -796.29
2 33,560 33,199.84 360.17
3 33,250 33,263.38 -13.38
4 34,670 33,326.93 1,343.07
5 32,340 33,390.48 -1,050.48
6 33,570 33,454.03 115.97
7 33,280 33,517.58 -237.58
8 34,670 33,581.13 1,088.87
9 32,700 33,644.68 -944.68
10 33,920 33,708.23 211.77
11 33,630 33,771.78 -141.78
12 35,030 33,835.33 1,194.68
13 33,340 33,898.87 -558.87
14 34,360 33,962.42 397.58
15 32,050 34,025.97 -1,975.97
16 35,470 34,089.52 1,380.48
17 33,540 34,153.07 -613.07
18 34,760 34,216.62 543.38
19 32,450 34,280.17 -1,830.17
20 35,870 34,343.72 1,526.28



THE MEAN SQUARE ERROR 997,111.23
THE FORECAST FOR PERIOD 21 34407.27


So i would need some help in interpreting so that i could give some suggestions about the future in terms of advertising, marketing etc.

Thanks,
Ecce Homo

Originally Posted by EcceHomo

Hi everybody,

I am doing the quantitative methods project. It is mainly concerned with the forecasts and decisions. The only data that i have got were sales for 20 different time periods. I have done different forcasting methods and now i have chosen the "best one" . The one with smallest mean square error. In my case it was a linear trend model. And i got the following data for the forcasted sales (also for the next period). From the data below i need to make a discussion of - the forecasts and suggestions about the future in terms of advertising, marketing etc. As i have already mentioned i have also made other forecasting methods (Moving averages 2, MA3 and MA4 , together with linear trend, multiplicative model) I have made a graphs for each model so that i can more easily try to make a suggestions for the future!

Here is the graph for the best forcasted method (linear trend)



*****************************

THE LINEAR TREND EQUATION:


T = 33072.737 + 63.549 t

where T = trend value of the time series in period t



TIME PERIOD TIME SERIES VALUE FORECAST FORECAST ERROR
=========== ================= ======== ==============

1 32,340 33,136.29 -796.29
2 33,560 33,199.84 360.17
3 33,250 33,263.38 -13.38
4 34,670 33,326.93 1,343.07
5 32,340 33,390.48 -1,050.48
6 33,570 33,454.03 115.97
7 33,280 33,517.58 -237.58
8 34,670 33,581.13 1,088.87
9 32,700 33,644.68 -944.68
10 33,920 33,708.23 211.77
11 33,630 33,771.78 -141.78
12 35,030 33,835.33 1,194.68
13 33,340 33,898.87 -558.87
14 34,360 33,962.42 397.58
15 32,050 34,025.97 -1,975.97
16 35,470 34,089.52 1,380.48
17 33,540 34,153.07 -613.07
18 34,760 34,216.62 543.38
19 32,450 34,280.17 -1,830.17
20 35,870 34,343.72 1,526.28



THE MEAN SQUARE ERROR 997,111.23
THE FORECAST FOR PERIOD 21 34407.27


So i would need some help in interpreting so that i could give some suggestions about the future in terms of advertising, marketing etc.

Thanks,
Ecce Homo

Are you sure that a model with a four period seasonality would not have a
smaller mean square error?

RonL