You are not currently logged in.
Access JSTOR through your library or other institution:
If You Use a Screen ReaderThis content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Daily Cash Forecasting and Seasonal Resolution: Alternative Models and Techniques for Using the Distribution Approach
Tom W. Miller and Bernell K. Stone
The Journal of Financial and Quantitative Analysis
Vol. 20, No. 3 (Sep., 1985), pp. 335-351
Published by: Cambridge University Press on behalf of the University of Washington School of Business Administration
Stable URL: http://www.jstor.org/stable/2331034
Page Count: 17
You can always find the topics here!Topics: Workdays, Analytical forecasting, Cash flow, Cash, Time series forecasting, Statistical estimation, Forecasting models, Forecasting techniques, Coefficients, Collinearity
Were these topics helpful?See something inaccurate? Let us know!
Select the topics that are inaccurate.
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Preview not available
Daily cash forecasting generally requires some method to reflect day-of-month and day-of-week effects. It requires the resolution of multiple seasonals, a problem given scant treatment in the econometrics literature. This paper first presents multiplicative and mixed-effects specifications of day-of-month and day-of-week effects as alternatives to the additive specifications. Then, several important estimation issues pertinent to each specification are investigated, namely collinearity, holiday effects, length-of-month distortion, varying weekly-monthly pattern mix, and daily-monthly consistency. The paper develops a broad class of distribution-based linear forecasting models in great generality similar to the way that Box and Jenkins  provide a broad class of time-series models that can be specialized via parameter selection (specification). In our case, parameter selection (specification) gives particular members of the linear class of distribution models. A particular version can be tested against an alternative specification via hypothesis tests on model parameters.
The Journal of Financial and Quantitative Analysis © 1985 University of Washington School of Business Administration