Forecast of Revenue Freight Carried by Rail in Texas to 1990

Access full-text files




Williams, David L.

Journal Title

Journal ISSN

Volume Title


Council for Advanced Transportation Studies


Revenue tons of freight carried by rail in Texas have been forecast to 1990 using multiple regression analysis and trend analysis. Data were gathered on the dependent variable (revenue tons of freight carried by rail in Texas) and on twenty-nine independent variables (economic indicators of the Texas economy) for the base period 1950 to 1972. Missing values from the time series data were estimated with the aid of the OMNITAB computer program POLYFIT. Multiple regression analyses were used to measure the linear relationship between the dependent variable and a set of independent variables, taking into consideration the interrelationships between the independent variables. From these analyses, a set of ten independent variables was selected as providing the best predictor regression equation. Forecasts for each of the ten selected variables were computed for 1975, 1980, 1985, and 1990 by extrapolating a chosen trend curve. These forecasted values were then substituted into the regression equation to yield forecasts for the tons of revenue freight carried by rail in Texas.


See this work in the Center for Transportation Research Library catalog:

LCSH Subject Headings