Roxyndallas,
The determination of the cost of capital is a forward looking concept, reflecting the rate of return that is needed to attract capital funds to the risky project on a going forward basis. Therefore, the use of historical ERPs and historical beta estimates is useful only to the extent that these estimates proxy for forward looking expectations of risk. Thus, using the beta reported on Yahoo Finance, or even the historical equity risk premium, may or may not be appropriate for Apple. You need to assess the facts and circumstances of your case to determine if Yahoo Finance's beta estimate is an appropriate forward looking estimate. The same idea applies to the equity risk premium and the capital structure components used in your determination of the WACC. Unfortunately, in valuation there is no cookie cutter approach. It's not as if Yahoo Finance's beta is the correct beta, because there are many different estimates of beta. The correct beta is the best beta that estimates forward looking risk. You need to use your reasonable judgment to assess whether Yahoo's beta is appropriate. Remember, betas are estimated using realized returns. So some common questions you need to ask include (a) does the historical period that Yahoo used to determine beta provide an unbiased estimate of beta, (b) do you need to consider a longer or shorter beta estimation period (c) has Apple experienced any significant fundamental change in its business model or is Apple expected to have any significant change in its business model such that a historical beta estimate is a poor measure of risk, (e) compute the standard error of the beta estimate of Apple to determine how statistically significant the factor loading is; you may be have no confidence in the estimate once the range is established, (f) what risk free rate does Yahoo Finance measure betas relative to; do you need to adjust the beta. As you can see there are many questions you need to ask.
Also, in terms of the selection of the risk free rate, if you are trying to determine the value of Apple, then the appropriate risk free rate is a 10 or 20 year treasury because the maturity on this type of bond most closely matches the cash flow duration of a business (i.e. a perpetuity).
Determining the equity risk premium is subject to similar issues. The selection of the ERP is not simply whatever is reported in Damadoran's schedule. That is only an estimate of the forward looking equity risk premium. It may be or may not be the best estimate. Generally, you have three or four options to determine the equity risk premium (a) examine the historical risk premium over a defined time period, (b) infer a forward looking estimate using the current level of the market index, cash flow growth, etc. (c) use market surveys, or (d) supply you own. Again, the equity risk premium should reflect your best estimate of what investors require to compensate for risk on a going forward basis. Generally, I prefer to estimate the equity risk premium using the index level and cash flow expectations, because this provides an actual estimate of the forward looking premium. I will tell you that historically (from 1926 through December 2009). the historical realized supply side premium has been about 5.25%. Since 1960ish through December 2009, the historical realized premium has been about 4.25%. The forward looking estimates in Damodaran's website (when adjusted relative to 20-year treasuries), has ranged from 4.5% to 5.00% in recent months. So, in my opinion, a good estimate of the forward looking premium is probably somewhere between 4.25% and 5.25%, say 4.75%.
Hope that helps