On the Stationarity of Savings in the United States

Authors

  • Noel Brodsky Eastern Illinois University

DOI:

https://doi.org/10.1016/1057-0810(95)90019-5

Abstract

The question of stationarity of saving as a percent of disposable personal income is examined in detail. National Income and Product Accounts data is taken annually from 1929-1988, and quarterly from 1970-1988. The initial approach uses Dickey-Fuller (1987) tests. Their results indicate that, depending on the time period chosen, one is led to believe that both stationarity and non-stationarity exist in the data. ARIMA models are then fitted to both annual and quarterly data. Relatively large non-difference models are necessary to fit the long-range annual data (1929-1988), while relatively small differentiated models fit quarterly and short-range annual (1946-I 988) models. It is concluded that at least for saving as a percent of disposable income in the US, it is reasonable to believe that saving is not a stationary process. Quarterly Review of Economics & Finance, Summer 1994, 34(2):

183-I 93. (Reprinted with permission of ABIfInform, Copyright UMI.)

Published

1995-06-30

How to Cite

Brodsky, N. (1995). On the Stationarity of Savings in the United States. Financial Services Review, 4(1), 61. https://doi.org/10.1016/1057-0810(95)90019-5

Issue

Section

Abstracts of Articles on Individual Financial Management