The purpose of this study was to investigate the stochastic properties of quarterly revenue and expense data. Firm-specific and cross-sectional models were determined for a sample of 55 firms' data. The study utilized the time-series methodology of Box and Jenkins. Evidence regarding the stochastic properties of accounting numbers impacts upon their interpretation and utilization in decision-making contexts. / The results of this empirical investigation suggest that the time-series properties of revenue and expenses are similar to those of earnings. Each series examined was characterized by non-stationarity in its original form, seasonality, and a quarter-to-quarter and quarter-by-quarter relationship. Sales were described by a (0,1,0) x (0,1,1) cross-sectional model, while the expense series were described by a (0,1,1) x (0,1,1) or Watts-Griffin model. In general, the cross-sectional models of revenue and expenses were superior relative to firm-specific models. An examination of quarterly earnings revealed that the Brown-Rozeff and Watts-Griffin models were the best predictors of quarterly and annual earnings. These results suggest that parsimonious models offer a promising approach to forecasting quarterly revenue, expense, and earnings data. / Source: Dissertation Abstracts International, Volume: 41-01, Section: A, page: 0313. / Thesis (D.B.A.)--The Florida State University, 1980.
Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_74020 |
Contributors | KEE, ROBERT CARL., The Florida State University |
Source Sets | Florida State University |
Detected Language | English |
Type | Text |
Format | 205 p. |
Rights | On campus use only. |
Relation | Dissertation Abstracts International |
Page generated in 0.0013 seconds