Business Forecasting Group Project
Course Project (14% + 3% in Total)
1. This project has a value of 14% of the total assessment. In addition, there is a teamwork component worth 3%. The teamwork mark will be based on the online self and peer assessment (see Teamwork Assessment section at the end of this document).
2. This project must be completed in a group of 3 or 4 students. The members of a group come from the same tutorial class. Groups have been alphabetically assigned. Each group is identified by a class number (e.g. Class 2649) and a group number (e.g. Group
2). Please check the spreadsheets in “courseProjectGroups” to see the group you belong to. Each group must select one person to submit the project. Each group …show more content…
At minimum, a thorough analysis in the framework of the classical decomposition is expected.
The data file, buildingApprovals2009Dec.xls, has two columns: Date and
Approvals. Note that Approvals are measured as the number of building approvals in New South Wales, and can be treated as real numbers (i.e. they are not restricted to integers).
A template EViews file, assign_temp.prg, is provided, which contains commands for reading data in EViews, generating time index and seasonal dummies. To carry out the computation of the project, it may be helpful to consult and borrow from the examples in the lectures (BF-09 in particular).
Use time polynomials up to the third order (at most) to model trend if any. Use monthly dummy variables to model seasonality if any.
Pay attention to the ACF and PACF of the cyclical component. Note that pure AR models, if suitable, are easier to estimate and analyse than MA or ARMA models.
In particular, the stability analysis for AR models can be done in EViews.
Use SIC (or AIC) to select your preferred model.
Testing for seasonality can be done in EViews by the commands equation eqn1.ls y c D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12 freeze eqn1.wald c(2)=c(3)=c(4)=c(5)=c(6)=c(7)=c(8)=c(9)=c(10)=c(11)=c(12)=0
where time trend (if any) or AR terms may also be included at the end of the equation. You may directly model Approvals (without logarithm).
The Course Project provides an environment for students to