Question

Consider the following quarterly time series.

​

Quarter	Year 1	Year 2	Year 3
1	923	1,112	1,243
2	1,056	1,156	1,301
3	1,124	1,124	1,254
4	992	1,078	1,198

​

Use a multiple regression model with dummy variables for quarters 1, 2, and 3 and a time variable. What is your forecast for year 4 quarter 1? Round your answer to one decimal place.

Answer 1

The forecasted value is 1317.9

The working the done using R, script can be found in the end.

The dummy variables are created and data has been organised as follows:

Q1 = 1 ; Quarter 1

0 ; otherwise

Similarly Q2 & Q3 are created.

Data:

R Script:
> model = lm( Value ~ Year+Q1+Q2+Q3, data = timeseries )
> model

Call:
lm(formula = Value ~ Year + Q1 + Q2 + Q3, data = timeseries)

Coefficients:
(Intercept) Year Q1 Q2 Q3  
864.083 112.625 3.333 81.667 78.000  

> newdata = data.frame(Year=4, Q1=1 , Q2=0, Q3=0)
> newdata
Year Q1 Q2 Q3
1 4 1 0 0
> predict(model,newdata)
1
1317.917
> round(predict(model,newdata),1)
1
1317.9