A multi-market model relating the variance of relative price changes to unanticipated inflation and real income was developed by Richard Parks for an article in the Journal of Political Economy. Parks used annual data in his estimation, and his use of the current rate of inflation as a predictor for next period's rate was rather simplistic. In this paper, his model was tested with two alternative specifications for anticipated inflation and with quarterly rather than annual data for the period 1947 through 1978. Anticipated inflation was estimated by (1) a time-series of past interest rates, and (2) a time-series of past inflation rates and the money supply. The multi-market model was estimated by employing the Cochrane-Orcutt iterative technique. The regression results gave additional support for Parks' model, but the respective roles for the two causal variables, unanticipated inflation, and real income, were reversed. Unanticipated inflation was seen to have a stronger effect in the quarterly data than it had in Parks' estimation with annual data. Relative price changes that result from an inflation that is unanticipated was said to be a temporary phenomenon. This was suggested to be the reason for the role reversal of the two causal variables because a temporary relationship such as the model attempts to estimate would be expected to show itself more significantly in quarterly data than it would in annual data. It was also suggested that unanticipated inflation may play a role in the persistence of staflation. / M.A.