Forecasting the Money Supply in Time Series Models

In this paper, time series techniques are used to forecast quarterly money supply levels. Results indicate that a bivariate model including an interest rate and M-1 predicts M-1 better than the univariate model using M-1 only and as well as a 5-variable model which adds prices, output, and credit.

The paper also presents evidence on the issue of using seasonally adjusted data in forecasting with time series models. The implications of these results apply to all econometric modeling. Results support the hypothesis that using seasonally adjusted data can lead to spurious correlation in multivariate models.