Show me the dividends: A review of the dividend discount model.

Today I am review the dividend discount models. This is the start of an extensive review on how the intrinsic value of a stock is estimated. The basic dividend discount model seems to be the most simplified variation. To the best of my understanding states that the stock price should equal the present value of all future dividend payments plus the final sales price (or perpetuity if you are never planning to sell.).

(Imagine an equation here as soon as I find away to add equations on blogger)

The recurring problem you will find with this and any of the valuation models is that you must accurately estimate future dividends, present value on a future sell date and the market capitalization rate. Any slight error in any three estimates can lead to a noticeable deviation from your calculated intrinsic value and the actual intrinsic value. Any determination on whether a stock is under valued or overpriced will be incorrect.

Equity Valuation Models: I have to start somewhere

So my plan was to jump head first into studying for my Series 24 but that has been delayed by situations outside of my control. But... I need to keep learning and growing. So I amended my plans. I will still keep toward my goals. Although this admittedly makes them more challenging (which I actually invite). After some soul searching and some analysis on where I stand knowledge wise I have added ANOTHER item to the list. Last year I studied Equity Valuation Models briefly. I throughly enjoyed learning about dividend discount models but after going over the items that previously studied I realized I still did not have a firm grasp with the concepts of valuation models. So I have decided to not only review everything I studied but also go more in depth with greater rigor on the subject. I feel like learning this will be invaluable and because I enjoyed it previously I hope the new academic rigor will reignite the passion. So with further ado I start with Book Value.

Back to the Basics: Setting goals for 2010

2010 is a big year for me personally with my upcoming marriage to my beautiful fiancee. So I have been racking my brain to map out ways to make it a big year professionally. So I set out some goals to help me improve my financial expertise. I wanted my goals to be SMART Specific, Measurable, Relevant, and Timely. So here they are.

1. Study, take and pass Series 24 by June

2. Study, take and pass Series 66 by December

2. Study, take and pass CFA Level 1 Exam by the end of the year

3. Write a fiction/finance book by December

4. Finish reading a book per month.

Now this seems daunting and some may say unrealistic. But I have always been one to say if I aim for excellence even if I fall short I will hit pretty damn good. 2010 is a year for growing both personally and career wise. I want to prove that I can achieve both.

Can skilled investors make consistent abnormal trading practices. If yes, how can we distinguish skill from luck?

Regression Analysis exposed

Single-variable regression model. I've had to go back at least three time to this concept because it seems to play a crucial role in having a in depth understanding of CAPM. Now again let me prefix by saying I know there are objections to the CAPM for risk analysis but that does not take away from the fact that it is commonly used and taught. And for that reason enough I could not claim to be a good equity analyst (I' can't even claim to be an equity analyst yet) unless I understand the fundamentals behind this theory and appropriately decided for myself the usefulness.

I regress. But I am using a website http://www.econinformation.com/cgi-bin/wea/regression_analysis.cgi to name a few to develop a understanding of regression analysis so that my foundation for understanding more complex models (multi-variable??) will be solid.

Here is what I know so far.

Regression analysis is the process of finding an equation that allows me to find the relationship between a variable that I wish to predict the value of (independent variable) and the variable (independent variable) that I believe can be used to find the values of the dependent variable. 

Fortunately better academics (for now?) have developed a equation that allows us to analyze the relationship between the return of an asset (let's stay stock) in excess of the risk free rate and the excess market return whose values I (academics) believe can be used to deteremine the excess return of the stock. Let's see the equation

     (thanks Wikipedia http://en.wikipedia.org/wiki/Single-index_model)

This looks very cryptic so let me explain the equation using the definition of single variable regression equation.

(Deep Breath) 

So let's let i equal the asset (let's say stock from now on).

Let t equal a certain time period.

Let ri equal the return of the stock i and rm equal the return of m (which is the market)

E is equal to the firm specific risk or the residual

a is the excess return considering the market value. This is called alpha and is what active management trys to maximixe. But according to CAPM this should be 0. 

B (or B looking character) is beta and represents the stocks sensitvity (responsiveness) to the market. 

Ri is the excess return of the stock i over the the risk free rate (remember we only care about excess because without excess return everone would just invest in risk free assets)

Rm is the excess return of the market over the risk free rate.

(Exhale)

So now Ri is dependent on Rm. I (academics) believe that given the values for the excess return of the market Rm we can find Ri if we know how the stock i moves in relation to the market. 

That's enough for now.

 

What is my most efficient combination of starbucks ingredients that will give me the optimal caffeine rush

I just finished reviewing how to develop the complete portfolio by determining the optimal risky portfolio and combining that with a risk free asset. So just so I can remeber it better let me deconstruct it into 3 steps.

Determining Effficient Diversification with Many Risky Assets

1, Determine the efficient frontier. - This requires finding the "best" investment opportunity set which is the set of portfolios that offer the highest reward to volatility ratio. (It's slightly more involved than that but this is a brief overview)

2. Choose the optimal Risky Portfolio - Now that we have the efficient frontier we want to find the optimal risky portfolio from the set of portfolios on the efficient frontier. To find the optimal risky portfolio we have to optimize the reward to volatility ratio. The optimization of the reward to volatility ratio involves the risk free asset. Using the current risk free rate search for the CAL with the highest reward to volatility ratio (i.e. the steepest slope). This will be the risky portfolio that will always be used by investors no matter there risk aversion because it offers the best reward to volatility ratio.

3. Choose the appropriate mix between the optimal risky portfolio and the risk free asset based on the risk aversion of the client. Once done you will have the complete portfolio.

So there you go. Very interesting, maybe slightly unrealistic and certainly doesn't fit into all situations but as I study more this basic concept seems to have been used to develop more rigourous theories that may better reflect real life scenarios. Next up Capital Asset Pricing Model (CAPM)  Yay.

Understanding the argument for "Time Diversification"

So my mind has started to understand the argument for "Time Diversification". My understanding reads like this. Diversification among many risk assets across many years will lower the standard deviation of the portfolio.

Let's break this down further. In simple terms stocks have 2 types of risk, systematic (market risk) and non-systematic risk (firm specific). Market risk is determined by the market movement and the beta of the stock (it's correlation to the movement of the market). This type of risk cannot be "diversified" away by increasing the number of securities in the market. Non-systematic risk or firm specific risk can be virutually elimanated by the increasing the number of stocks. This is because each security's market risk is dependent on the market. This is not the same for firm specific risk because this risk is unique to each security. For example Google and Mcdonalds both have market risk as a part of there overall risk. How the market risk effects there return is determined by the beta of Google or Mcdonalds. But a firm specific risk in Google has no effect on the firm specific risk of Mcdonalds. Because firm specifc risk are independent as the number of securties increase the risk will offset each other to the point where one problem security will not effect a portfolio of 10,000 considering all are equally weghted. A clarification is needed in that firm-specific risk example can extend to sector specific as well.

So now that we know we can effecively eliminate firm(sector) specific risk by diversifying across many assets and sectors and the main reason this is possible is because the firm specific risk or the sector specific risk are independent of each firm and sector. This independence allows for offestting of "bad" returns.

Well this idea can be extended to time. Right? Because stock returns in successive years are almost uncorrelated an increase in years will allow for offeting of bad years with good years. Sounds reasonable right?

Well it's time for me to start reading why this logic could be flawed.