Which method is practically better?

If in an ideal scenario, a person, who is calculating the cost of equity, has all the information about dividends, growth, risk premiums, systematic risk. He can go with any of the 3 different ways (3 included in the level 1 curriculum) of calculating the cost of equity (Dividend discount method, CAPM, Bond Yield and Risk Premium). What should be his best choice?

I understand that all the methods have their own assumptions. I just want to know which of these options the best in real life scenarios is. What methods do analysts prefer? Which one presents the most realistic result?

Its a tough one between DDM and CAPM. As you have mentioned, they all have different underpining assumptions.

I am struggling to say one is better than the other.

For me, i will use just any of the two.

Maybe some more experienced people can share more light on this.

The problem is that such an ideal scenario does not exist. Sometimes you lack of data, the market is not efficient, or transaction costs can distort a given model results (violate assumptions).

I would use dividend discount method on a very mature company, with steady dividends and a well-known long-term growth rate (for mature companies we can assume the growth rate of the company is reflected in the EPS growth rate)

I would use the CAPM model when you be able to gather the relevant data for the model: risk free rate with the correct maturity, unlevered and levered beta that better suits the company, and an adequate market return benchmark.

For the last one, you need to be careful on what the risk premium will be, you need another model to calculate it… You may use benchmark on this if the market is efficient. If not, the task complicates.

On the Bond Yield+Risk Premium, 3-5% risk premium is used as a rule of thumb.

If the information is available CAPM is the industry standard either calculated directly or via a peer group.

For the several components of the CAPM model (RFR, beta,…) there are different sub-methods (like Svennson method for the RFR or the Vasicek method for calculating betas). I’m not sure but I think that these methods go beyond the CFA curriculum.

Other methods should be used for plausibility.

Best, ​Oscar