the loss of degrees of freedom when additional independents variables are added to a regression.
that the error term in the regression is covariance stationary. Therefore the two time series are cointegrated.
The parameters and standard errors from linear regression will be consistent and will allow testing of the hypotheses about the longterm relationships between the two series.
Structural Model  Reduced Form Model 



TAX
and is defined as operating profit divided by capital employed
ROCE is useful when comparing peer companies in different countries because the comparison of underlying profitability would not be skewed by low taxes.
Capitalization Rate = WACC – Long term Growth Rate
EBITDA overestimates cashflow from operations if the company’s working capital is growing.
Lack of Control Discount = 1 – (1 / (1 + control premium))
Remember: Discounts are usually multiplicative rather than additive.
CEY = CBY – b x LTEG + Residual
CEY = current earnings yield
CBY = Moody’s Arated corporate bond yield
LTEG = consensus fiveyear earnings growth
b = coefficient measures the weight the market gives to fiveyear earnings projections
Justified P/E = 1 / CBY – b * LTEG
 It does not consider an equity risk premium.
 It ignores the portion of equity value that comes from longterm earnings growth.
 It compares a real variable (index yield) to nominal variable (government bond yield) which makes it unhelpful when inflation is either nonexistent or very high.