The Effects of Data Aggregation in Statistical Analysis

The problem of ecological correlation is now widely recognized but detailed analyses of the effects of aggregation on correlation and regression coefficients are rare. A short review of the aggregation problem is followed by an analysis of the specific effect of proximity aggregation on the slope coefficient of a bivariate linear model using data drawn from the Los Angeles Metropolitan region. The evidence suggests that changes in the slope coefficient are best related to the manner in which the covariation between the independent and dependent variables changes with increased aggregation.     

Dimensions of Spatial Price Dispersion Before and After the Crash of 1929

The spatial and temporal patterns of prices are analyzed over the 1919–40 period. Arguments are made for the distinctiveness of different economic episodes, including the crash of 1929, in terms of their local price effects. Nevertheless, it is also shown that the variability between cities and between components of their price series declined through the early 1940s. Alternative conceptions of spatial price systems are introduced, and due regard is also given to the structure and interpretation of city CPI data. Implications are drawn regarding the proper design of a theory that would explain the observed patterns of spatial price dispersion. This larger project is left, however, to another time.     

HAVE CHANGES IN LOCALIZATION ECONOMIES BEEN RESPONSIBLE FOR DECLINING PRODUCTIVITY ADVANTAGES IN LARGE CITIES?*

This paper investigates whether economies of urbanization and localization changed in two-digit manufacturing industries over the period 1967-1977. Estimates of changes in such economies indicate that the localization advantages of large industries declined in several industries, but that urbanization advantages increased for a small number of industries and decreased for a small number. Moreover, the absolute change in two-digit industry employment tends to be negatively correlated with industry size, city size held constant, and positively associated with city size, industry size held constant. These latter findings tend to support the findings regarding changes in external economies.     

TRENDS AND CYCLES IN THE COMPOSITION OF THE FEDERAL BUDGET, 1947–48

In this study, systematic movements in the composition of the Federal budget are studied, using a simple statistical structure which distinguishes trend, cyclical, and price-indexing phenomena. Using NIA data on budget expenditures and receipts, it is found that share movements on both sides of the budget are dominated by trends. Among expenditure categories, the shares of transfer programs show strong upward trends while the defense purchases share has beentrending strongly downward. On the revenue side, the shares represented by social insurance contributions and personal income taxes have been trending upward at the expense of corporateincome taxes and indirect taxes. Significant cyclical and price-indexing responsiveness is shown by several expenditure and revenue categories. A noteworthy finding is that the estimates show little evidence of structural change over 1947-1978, suggesting that the trend and otherforces found to be significant are not of recent origin but have been at work over the whole period.