Forecasting annual train boardings in Melbourne using time series data

Abstract

This paper analyses the relationship between yearly train boardings in the Melbourne metropolitan area and six theoretically explanatory economic and demographic variables over the period twenty seven year period 1983-84 to 2009-10. Two of these variables were lagged by three months, and by six months. A series of univariate linear regression analyses was performed, followed by several multivariate regression analyses of various combinations of the independent variables which showed the highest explanatory power. The three independent variables with the highest explanatory power are in descending order the annual average percentage of total interest payments to household income(lagged 3 months), the estimated resident population in the Melbourne Statistical Division (X6), and the estimated number of persons employed (both full and part time) in the Melbourne Statistical Division. Multivariate regression analyses yield several usable forecasting equations. The one of which is the basis of the presented forecast of train patronage from 2010-11 to 2012-13 comprises three explanatory variables: the real average annual price of a zone 1 ticket, the real average annual price of unleaded petrol (lagged 3 months) and the estimated resident population of Melbourne. Time series forecast are also presented. Both types of forecasts indicate an increase in train patronage from 2010-11 to 2012-13 that is double the average annual rate of growth experienced in the previous three years. Limitations of this research and future research plans are outlined

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 8 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.