Question

Consider the following time series data. Month 1 2 3 4 5 6 7 Value 23 11 21 10 20 23 15 Round your answers to two decimal places. a. Compute MSE using the most recent value as the forecast for the next period. Mean squared error is What is the forecast for month ? b. Compute MSE using the average of all data available as the forecast for the next period. Mean squared error is What is the forecast for month ? c. Which method appears to provide the better forecast?

Answer 1

Month	1	2	3	4	5	6	7
Value	23	11	21	10	20	23	15

a. Compute MSE using the most recent value as the forecast for the next period.

recent value method includes the forecasted value from the most recent period

Eg. Current month's forecast value = Last month's actual value

Month	1	2	3	4	5	6	7	8
Value	23	11	21	10	20	23	15	-
Forecast	-	23	11	21	10	20	23
Error (Forecast - Value)		12	-10	11	-10	-3	8	-
		144	100	121	100	9	64	-

Mean squared error is = = ...........We use 6 since we have 6 set of forecast with their actual values

=

What is the forecast for month 8? =

b. Compute MSE using the average of all data available as the forecast for the next period.

Month	1	2	3	4	5	6	7	8
Value	23	11	21	10	20	23	15
Forecast	-	23.00	17.00	18.33	16.25	17.00	18.00
Error		12.00	-4.00	8.33	-3.75	-6.00	3.00
		144.00	16.00	69.44	14.06	36.00	9.00

Average =

Eg For forecast of 5 th month we will take sum of values upto the month of 5 (23 + 11 + 21 + 10)

= 65

Average forecast for 5th month = 65 / 4 = 16.25

Mean squared error is = =

=

What is the forecast for month 8? =

c. Which method appears to provide the better forecast?

Average forecast

This is because it has a lower MSE (lesser errors ) than Recent value forecast method.