Since Markowitz [15, pp. 98–101] and Sharpe [19] developed the single-index market model (SIMM hereafter) it has received considerable research attention. Empirical tests have established the model's econometric significance [3, 12, 13] in partial equilibrim analysis. However, research into the relationship between the SIMM and its macroeconomic environment has been meager. It has been shown that the market factor changes intertemporally [13, 16, 18, 20]. However, whether these changes in the market factor and, more basically, changes in the macroeconomic situation affect the SIMM is unknown.

After Markowitz [14, p. 100] and Sharpe [19, 20] suggested estimating the beta systematic risk coefficient for market assets, finance professors, stock brokers, investment managers, and others began expending large quantities of resources each year on estimating betas. Unfortunately however, it appears that the ordinary least-squares (OLS) regressions used in nearly every instance may be inappropriate. This paper suggests that many stocks' beta coefficients move randomly through time rather than remain stable as the OLS model presumes.

In this paper, we find significantly higher preholiday returns in futures contracts compared to nonholiday returns. The findings are consistent with the inventory adjustment hypothesis, since higher preholiday returns associated with lower trading volume are most pronounced for exchange‐closed holidays. There is evidence of positive postholiday returns associated with higher trading volume for exchange‐open holidays. This is consistent with positive holiday sentiments. The holiday effect is uniquely independent: The magnitude of excess holiday returns is the largest among all seasonal variations.

Based on the theory of the pricing of capital assets developed by Sharpe [12], Lintner [9] and Mossin [11], Professor Jensen formulated a return-generating model to measure portfolio performance [5]. In a subsequent paper, Professor Jensen [6] investigated the impact of the investment horizon on the functional form of the model. Lee [8] has proposed a generalized specification of the model to resolve this problem. Alternative estimation methods for testing the linearity of the model in terms of time-series data have also been suggested by Lee. Moreover, the stability of the beta coefficient over time and the impact of the market's condition on both the alpha (or, Jensen's measure of performance [5]) and beta of the model have come under scrutiny in financial research.

Monthly returns are used to estimate the single-index market model (SIMM). Binary variables are used to determine if the alpha intercept and beta slope coefficients are stable through alternating bull markets and bear markets. The results suggest that some investment analysts have fallen into the trap of misapplying econometric models and, as a result, are purveying erroneous information. Neither the alpha nor the beta statistics in the SIMM appear to be significantly affected by the alternating forces of bull and bear markets. Of course, the SIMM, the alpha, the beta and the statistics from all econometric models change from sample to sample. But, the question addressed here was whether or not these normal sampling errors were more than would occur in the classic stability tests. Such instability would tend to depreciate the value of the received risk-return theories. However, the SIMM was found to be unaffected by the three different bull and bear market conditions which were delineated.