Demand Forecasting

ASSIGNMENT 1: DEMAND FORECASTING

Assignment Tasks

The spread sheet on the last page of this document (see Attachment 2) contains data about:

•      The number of boardings (in millions) of passengers onto Melbourne trams over

 the years 1983-4 to 2011-2012. This variable is the dependent variable, and is denoted as Y_TRAM.

•      Data for the same number of years on eight other variables (denoted X_1,  X_2, X_2b X₃  …X_6.)  that theoretically or conceptually are associated with the dependent variable, and that might be relevant in forecasting future values of the number of tram boardings in Melbourne into the foreseeable future.

You should down load an electronic copy of this spread sheet (ASSIGNMENT 1 2013 DATA SET.xls) from the Assignments section of the OMGT1058 Blackboard. You will be required to use the Microsoft Excel package and specifically the data analysis section for this assignment. You are strongly advised to closely read UNIT 2 pages 21-58 before and during your work on the specific Tasks detailed below.

Specific required Tasks

Task 1:  Input the spread sheet into Excel.

Task 2: Make charts and briefly comment on any key trends (i.e. movement of data over time) in all the variables in the model.

Task 3: Construct charts of the relationship between the dependent variable and each of  the independent variables. Observe and briefly comment on the empirical

              relationship between dependent and independent variables depicted in these charts.

Task 4:   Run the linear regression function in Excel separately for each independent variable in the model

Task 5:  Analyse the regression statistics, ANOVA statistics and specific statistics about the coefficients values, standard errors and other diagnostic  statistics (i.e. t Statistic and

              P-Value) of the intercept and independent variables

Task 6:  Run a number of linear multiple regression models using the same independent  variables as in task 5. One way to start is to say the two or three independent variables Task 5 that demonstrated the highest explanatory features.  You need to try other combinations of independent variables as well.  Repeat Task 5.

Task7: Investigate whether competing mathematical forms improve the explanatory  power of the regression model used in Task 6 .

Task 8: Identify the most statistically and theoretically sound regression equation from those generated in Tasks 4 to 7. Justify your choice of a single predictive equation using the material contained in UNIT 2.

Task 9: Forecast the values of the independent variables for the model chosen in Task 8 for the years 2012-13 to 2014-15 inclusive). Briefly justify the method used and

the results of these forecasts.

Task 10: Forecast tram boardings (millions) for Years 30-32 i.e. 2012-13 to 2014-15 inclusive) using the forecast values of independent variables from Task 9 and the predictive equation for the selected equation from Task 8.

Task 11: Prepare an alternative forecast of tram boardings or the years 2012-13 to 2014-15 inclusive using a time series method. Explain your choice of the specific

              time series trend line  (i.e. exponential, linear, or logarithmic, or polynomial, or power, or moving average).

Task 12: Compare the forecast obtained from Task 10 with that from Task 11. Which one of the two would you be most confident in recommending for use by

            transport planners and decision makers? Why?

Task 13: Identify two other independent variables that are not included in the data set that you consider as candidate explanatory variables when it comes to

estimating Melbourne's tram demand, as measured by millions of boardings per year.  Justify your choice for both and specify the directional relationship that

economic theory suggests both of these variables coefficients should display.

Written report requirements

The main outcomes of Tasks 2 – 13 should be presented and discussed in a report format.  The report should be around 1500 – 2000 words (excluding figures and diagrams etc.) in total.

Do not include in the written report detailed excerpts from your Excel spread sheets. If included at all they should be contained in an appendix or a series of appendices.

The report should contain a brief (one page) executive summary that presents the main findings.