33、ontinue changing in the same direction. Blume (1971) further suggests that forecasts can be improved by adjusting the forecast beta so that it has the same mean as the historical mean.,37,The proposed adjusted beta estimate model is as follows: = 0.343 + 0.677 (average unadjusted beta based on data。

34、for the past three years) Some other studies propose to use multi-factor models to predict betas, as these factors are observed statistically significant affecting values of beta.,38,For example, Rosenberg and Guy (1976) consider the predictive power of other financial variables in forecasting beta, 。

35、 thereby estimating the following equation: Current = a + b1 (Past) + where bi represents a specific firm financial characteristic and n represents the number of financial characteristics used to predict beta.,39,They found that variables which help to predict betas include: 1. Variance of earnings. 。

36、 2. Variance of cash flow. 3. Growth in earnings per share. 4. Market capitalization (firm size). 5. Dividend yield. 6. Debt-to-asset ratio.,40,Rosenberg and Guy also found that even after controlling for a firms financial characteristics, industry group helps to predict beta. See Table 10.3.,41,Tab 。

37、le 10.3Industry betas and adjustment factors,42,SUMMARY,1. 单一指数模型将资产的风险来源分为系统性（宏观经济）因素或企业特有（微观经济)因素 。
指数模型假定宏观因素由证券收益的市场指数代表 。
2. 单一指数模型极大地减少了Markowitz组合选择过程中所要估计的组合收益与方差的参数个数 。
,43,3. 根据指数模型 ， 市场指数的超额收益与投资组合或证券之间的关系是：Ri = i + iRM + ei 投资组合或证券的系统性风险为i2M2, 两个资产i 和 j之间的协方差为ij2M 4. 对市场指数与证券或组合进行回归分析可以得出指数模型 。
该模型可以让我们检查特定证券或组合的收益如何根据市场收益发生变化 。
,44,5. 所估计的回归方程就是证券市场线 (SCL). 即, Ri = i + iRM. 证券市场线表示证券的具体收益是市场收益的函数 。
回归线的斜率是资产的贝塔系数 ， 而截距是样本期间资产的阿尔法值 。
6. 回归方程中的贝塔值等于CAPM的贝塔 ， 二者的区别在于回归中运用实际收益而CAPM 采用资产的预期收益 。
7. 随着时间过去 ， 贝塔值显示出回归值为1的趋势 。

来源：(未知)

【学习资料】网址：/a/2021/0413/0021924359.html

标题：单一|单一指数模型( 三 )