Wednesday, June 28, 2006

Girl...you will be a woman .... Soon....

Recently I watchedd Pulp Fiction ! This song was played when Uma Thurman returns from a party and dances ....

I really like this one. Sorry I did not acknowledge the original when I posted...

Girl, you'll be a woman... soon

I love you so much, can't count all the ways
I've died for you girl and all they can say is
"He's not your kind"
They never get tired of putting me down
And I'll never know when I come around
What I'm gonna find
Don't let them make up your mind.
Don't you know...

Girl, you'll be a woman soon,
Please, come take my hand
Girl, you'll be a woman soon,
Soon, you'll need a man

I've been misunderstood for all of my life
But what they're saying girl it cuts like a knife
"The boy's no good"
Well I've finally found what I'm a looking for
But if they get their chance they'll end it for sure
Surely would
Baby I've done all I could
Now it's up to you...

Girl, you'll be a woman soon,
Please, come take my hand
Girl, you'll be a woman soon,
Soon, you'll need a man

Girl, you'll be a woman soon,
Please, come take my hand
Girl, you'll be a woman soon,
Soon but soon, you'll need a man

Saturday, June 17, 2006

Anish !! We will miss you dude !

Socha tha sailing karenge. Kishti lekar lake ko paar karenge, kya pata thha kishti beech me jakar doob hi jayegi.

Congratulations to Anish Khandekar on completing his PhD. Now he is a Doctor !
Yesterday was his penultimate day in Madison and I guess we could make it memorable for him. 6 of us( manan, anand, jugal, santosh, anish and I ) left from our offices to the lake thinking we would sail and have beer in the middle of the lake. Manan was the crew leader. He is the only sailer amongst us.

Taking the sail boat we started to sail. The wind was nice, everyone was enjoying. In a few minutes, a heavier gust of wind came and the boat started to tilt. We tried hard to control it, but by the time we could realize it was not possible for us to control, three of us found ourselves in water and in next moment all 6 were in water. The boat tilted so much that the sail was in water. Everyone was holding the boat. Everyone panicked for a while but with the life jakets on, there was no problem in swimming. We started to enjoy the cold water of the lake.

The instant everyone was in lake, I was thinking we lost our beer, but it was floating. I collected as many cans as I could and Anish collected the rest. Anish was really enjoying. He opened a can and started to drink. I opened one can too. In a few moments the rescue team came and took us to the land.

Later Anish took us out for a great dinner and then to the bar.

Dear Anish, congratulations on completing your PhD and best of luck for a successful career.
You were a great friend here and we all will miss you !

Come back to Madison whenever you need to chill !!
~ Rohit.

Sunday, June 11, 2006

Stock Valuation

Stock Valuation

Stocks are infinitely lived assets

They pay dividends

The residual claim on the firm i.e. stockholders get dividends after bondholders get coupons.

Dividends are discretionary

To value the stocks, we estimate the present value of expected dividends for the life of the firm.

A closely related concept is the free cash flow to equity valuation model.

Free cash flow is defined as the cash flow generated by the firm that could have paid out as a dividend after funding required capital expenditures and making required payments to other claimants.

A group of models exists that are collectively referred to as DCF models. Theoretically, all of these models should provide an identical estimate for the stock price if properly estimated.

Present value computations

Consider an equity of $d per period. We know how to discount each period back by calculating the price of a zero coupon bond. So, the present value of an n period annuity can be calculated by summing the discounted cash flows. Recall that earlier we developed a formula for computing the value of an n period annuity.

A = a * ( z * ( 1 - zn ) / ( 1 – z ) )

For a stock we need to make three modifications.

Infinite life

Discounting a dividend cash flow which is a decision variable, not a coupon cash flow.

Stocks are more risky than bonds- default and residual claimant.

So a general formula for pricing the stock will be

P = d1z1 + d2z2 + d3z3

Z1 = 1 / ( 1 + k ) , z2 = 1 / ( 1 + k ) 2 and so on…

We must be careful using k instead of r.

K is market capitalization rate.

Simplifying assumption

Dividends grow at a constant rate for ever

P = do ( 1 + g ) / ( 1 + k ) + do ( 1 + g )2 / ( 1 + k )2 .

Take x = ( 1 + g ) / ( 1 + k )

We get P = do * x / ( 1 – x )

Substitute and get P = do( 1 + g ) / ( k – g ) = d1 / ( k – g )

K > g otherwise this stock has greater returns than the required rate of return.

If not, this would drive the price of the stock up and the price of the other stocks down.

Price earning ratios and constant growth

Assumptions

Dividend payout ratio is constant

dt = a * E

Where E is earning in period t and a is the payout ratio. So the firm pays out a constant proportion of their earnings as dividends.

Earnings grow at a constant rate g forever

Starting with the constant growth model

P = d1 / ( k – g )

Substituting in for d1 = a * E

P = a * E / ( k – g )

Now we can write the price earning ratio of the form as a function of the dividend payout ratio, growth rate in earnings and the required rate or return.

Divide both sides by E1

P / E1 = a / ( k – g ) = ( d/ E ) / ( k – g )

Limitations of constant dividend growth model

*Unlikely that the firm dividends grow at a constant rate. Most firms have rapid growth at beginning followed by normal growth

*Some firms pay no dividends followed by a policy of rapid dividends.

*K may be time dependent. That is a firm may change their business mix.

According to standard dividend discount model, a theoretical stock price is computed by discounting the stream of all future dividends at the market rates.

P = D1/( 1 + k ) + D2/ ( 1 + k)2 + D3/(1 + k )3

Where D1 = dividend at time 1

If dividends grow at constant rates g

P = D1 / ( k – g )

E1 = r Bi-1

Beta = ( 1 – alpha )

Bi = beta ( 1 + beta r ) I Bo

Ei = r( 1 + beta r ) i-1 Bo

Since Di = alpha r ( 1 + beta r ) i-1 Bo

Di = ( 1 + beta r ) i-1 D1

.g = Beta r = ( 1 – payout ratio ) ROE

P = alpha E1 / ( k – g )

Theoretical P / E = alpha / ( k – g )

P / E = ( 1 – beta ) / ( k – r beta )

P / E = ( 1 / k ) * ( 1 + beta ( r – k ) / ( k – r beta ) )

P = E / k + (r – k ) * ( g ) * ( B o ) / ( (k) * ( k – g ) )

Present value of growth opportunities

Pso = E1/ k + PVGO

An,k = E1 ( 1 – ( 1 + k ) –n ) / ( k )

In erms of formula

Pso = E1/ k + PVGO

E1/ P = k ( 1 – PVGO / P)

PVGO = ( gP – ( 1 – a ) E1 ) / k

Issues in valuation

Dividend discount in real terms

Vo = d1 / ( k – g )

We can separate real parts and inflation

Vo* = d1* / ( k* - g* )

Dividend forecasting and macroeconomic information

A structured way of pricing the dividends is as follows

First project the real GNP of a particular country for a number of years.

Second, determine the historical relationship between a company’s real assets and real GNP.

Project earnings given your GNP forecasts.

Third make an assumption about payment policies of the firms that you are examining. You can determine what the expected dividends will look like. Discount these back to the present – using the real required return for the firm.

Forecasting holding period returns.

It is quite difficult to forecast holding period returns for any investment. However these are some ways to get approximation to the holding period return. We will examine some simple ways to forecast the holding period return for the stocks and the bonds.

Stocks

A simplified approach will invariably make simple assumptions.

V1 = d2/ ( k – g ) = d1*( 1 + g ) / ( k – g ) = d1 * ( 1 + g ) / ( k – g )

V1 = Vo * ( 1 + g )

HPRcg = ( V1 – Vo ) / Vo = ( Vo * ( 1 + g ) – Vo ) / Vo = g

HPR = HPRcg + d1/ Vo = g + d1 / Vo

Bonds

Forecast of holding period return on bond will be linked to our estimate of what interest rates looked like in one year.

First the modified duration of the bond must be estimated. Second, make a forecast of interest rates.

HPRcg = - Modified Duration * Predicted Rate changes.

Finally, we need to allow for the expected coupon payment over the year that we are holding the bond.

HPR = HPRcg + c / B

So the holding period return is the sum of holding period capital gain plus the coupon yield.

Wednesday, June 07, 2006

Analysis of yield data.

Analysis of yield data

Yield curve usually tends to be upward sloping

Yields of various maturities and changes in these yields are highly correlated but this correlation decreases as the differences in maturity increases.

While there is a less similar dispersion around the mean yield change, for a longer bill, the price volatility is considerably greater than the longer term bills.

Bond Price movement and Duration

As interest rate goes up, the bond price decreases. A measure of volatility of bond prices is valuable. While a bond’s price maturity is a measure of interest rate sensitivity, it is not completely adequate since it only takes into account the final payment.

B = Sum ( Ct ( 1 + r ) –t )

The first derivative is dB / dr = - Sum ( Ct ( 1 + r ) –t – 1 t )

Adjust this by B and ( 1 + r ) as

D = - Sum ( Ct( 1 + r ) –t / B) t

It gives bond’s effective maturity by weighting the payments by their proportion of the bon’s present value.

It correctly measures the bond’s interest rate sensitivity to local changes in interest rates.

dB / B = - D * ( dr / ( 1 + r ) )

Rules of duration

Duration of a zero coupon bond is time to maturity

Holding the maturity and yield constant, the longer time to maturity increases duration

Duration of level perpetuity is ( 1 + r ) / r

Modified duration

= - D / ( 1 + r )

Generalized duration

dP / P = - D2 * d( 1 + r1 ) / ( 1 + r1 )

Convexity

This measures the degree of inward curvature of the yield to price curve. It’s the second order term of the Taylor series expansion of the change in Price to change in interest rates.

Co = - Sum ( D ( t + 1 ) / 2 )

The change in price over price is

d P / P = - D *( dr / ( 1 + r ) ) + Co * ( dr / ( 1 + r ) ) 2

An application of duration

Immunization

Its adjustment of the portfolio so that the risk due to interest change is minimized. An example is duration matching, convexity matching, and asset matching.

Bond Investment strategies

Passive bond strategies

Laddered- each year, simply reinvest the proceeds from the maturity bond.

Barbell – Heavier investments made in the short and long maturities. The maturing bond and a portion of 16 year bond is sold and reinvested in 5 year and 20 year bonds.

Index – A particular bond index is selected as a benchmark and a bond portfolio is bought to mimic the bond index. Given bond indices typically have thousands of bonds and given the lower liquidity in the bond market, most bond index investors use a sampling approach to implement this strategy.

Active investment- Portfolio is constructed to outperform the benchmark by using an investment strategy based on manager’s expectations of market events.

Active bond strategies

Issue selection

Sector selection

Interest rate selection

Directional

Yield curve positioning

Types of yield curve shifts.

Parallel shift

Twists

Butterflies

Sunday, June 04, 2006

Valuing Cash flows...

Valuing cash flows using zeros

Zeros are zero coupon bonds, which pay off at the maturity at a specified price.

Any portfolio of Bonds can be represented as a collection of ZCBs some of which will pay after one year, some will pay after 2 some will pay after 3 and so on.

Such a portfolio is called a replicating portfolio. The cost of replicating portfolio represents the price of the BOND portfolio. If the price of the replicating portfolio is more than its market price, then buy it and if it is less, then sell it ! The difference will be called as the arbitrage.

In general, arbitrage is the opportunity to make money by just buying and selling. Whether buying at lower cost and selling at a higher first or selling at higher first and buying at lower later doesn’t change the meaning of Arbitrage.

Wikipedia definition : In economics, arbitrage is the practice of taking advantage of a state of imbalance between two or more markets: a combination of matching deals are struck that capitalize upon the imbalance, the profit being the difference between the market prices. When used by academics, an arbitrage is a transaction that involves no negative cash flow at any probabilistic or temporal state and a positive cash flow in at least one state. A person who engages in arbitrage is called an arbitrageur. The term is mainly applied to trading in financial instruments, such as bonds, stocks, derivatives and currencies. Entrepreneurs seek out arbitrage markets (i.e. insurance) to earn a profit. For example, consumers benefit greatly from entities who locate arbitrage markets within the insurance industry. Insurance arbitrage markets are an immensely profitable trend that Wall Street is eyeing. Those insurance firms in a non-arbitrage environment will likely have difficulty competing against an arbitrage business model.

Law of one price: Portfolios with equal payoffs have equal prices.

Forward interest rate = TfT+1 = (MT+1 /M T – 1 ) = ( dT / dT+1 – 1 )

MT is what 1 dollar will become in T time.

dT is price of a zero coupon Bond.

Buying a bond for one year and forwarding it to three years, or buying it for 4 years, and similarly all types of combinations will give same payoff.

TfT+n = ((MT+1 /M T)n – 1 ) = ( (dT / dT+1)n – 1 )

From T year to T + n years.

Discount function is given by

dT = 1 / (1 + 0f1 ) *( 1+ 1f2 ) ….(1 + T-1 f T)

yT = ((1 + 0f1 ) *( 1+ 1f2 ) ….(1 + T-1 f T)) T – 1

yT = (1 / dT )T – 1

nfn+T = [(1 + Y n + T )n+T / (1 + Y n )n]1/T – 1

= [d(n) / d(n + T )]1/T – 1

d(t) = 1 / ( 1 + yT) T

General Bond Valuation

C = stated annual coupon rate of bonds

m = # of coupon payments per year.

c = periodic coupon actually paid

R = Stated rate

i = effective period rate.

k = years to maturity

n = number of coupon payments.

An i Value of annuity for n years at i interest.

Pz1= price of a zero coupon bond maturing in 1 year.

P b = c * Ani + m*Zn

BOND A

BOND B

R = Current stated interest rate

12.500%

12.500%

C = stated annual coupon rate of bonds

8.750%

12.625%

m = # of coupon payments per year.

2.000

2.000

c = periodic coupon actually paid

0.044

0.063

i = effective period rate.

0.063

0.063

k = years to maturity

12.0

12.0

n = number of coupon payments.

24.0

24.0

An i Value of annuity for n years at i interest.

12.266

12.266

Pz1= price of a zero coupon bond maturing in 1 year.

0.941

0.941

Bond Price should be

77.002

100.767

PERIOD

SPOT RATE

FORWARD RATES

DISCOUNT FUNCTION

STATED COUPON PAYMENT

STATED MATURITY PAYMENT

STATED BOND CASH FLOW

BOND PRICE

YIELD TO RETURN

1

8.15%

8.15%

0.92464

100

100

92.46

8.39%

92.26

0

2

8.20%

8.25%

0.85417

100

100

85.42

8.39%

85.12

0

3

8.35%

8.65%

0.78616

100

100

78.62

8.39%

78.53

0

4

8.40%

8.55%

0.72424

100

100

72.42

8.39%

72.45

0

5

8.40%

8.40%

0.66812

100

100

66.81

8.39%

66.84

0

6

8.40%

8.50%

0.61635

100

100

61.63

8.39%

61.67

0

7

8.40%

8.40%

0.56858

100

100

56.86

8.39%

56.9

0

8

8.40%

8.30%

0.52452

100

1000

1100

576.98

8.39%

52.49

524.9

9

8.40%

8.50%

0.48388

0

0.00

8.39%

0

0

10

8.40%

8.60%

0.44638

0

0.00

8.39%

0

0

1000

1091.2

1091

Estimating forward rates

Observe Price for a large number of coupon paying bonds

Denote Cash Flow as CFit

Dt be discount coefficients

P1 = d1 CF11 + d2 CF12 + ….+ e1

P2 = d1 CF21 + d2 CF22 + … + e2

Pn = d1 CFn1 + …

Yield to maturity

Yield to maturity

Calculate yield to return on the any call date. The lowest yield to call is known as yield to worst.

Bond Price Dynamics

Any bond can be brought to its Present value.

Vo = FV * ( 1 + R ) –T

dV/ dR = - T * ( 1 + R )-T-1* FV

Derivatives are negative – Price of zero coupon bond decreases with interest rates.

T affects the rate of decrease.

Measuring Bond Performance

Holding period return

HPR = ( Ct,t+1 +(Pt+1 - Pt)) / Pt

Time shape of cash flows

Price of a bond is dependent on the cash flows. Thus if the payoffs are concentrated in the beginning, then it will not be sensitive to changes in interest rates.

Graph of bond prices for different bonds show different curves

Term structure of interest rates and yield curves

Relationship between interest rates and time to maturity

These curves can be flat, monotonically increasing or decreasing or humped.

Term structure Theories

There are four term structure theories

  1. Market Segmentation Theory

Separate forces of supply and demand for funds in the short term Vs long term markets, with interest rates determined independently in each segment.

Investors have a strong preference for the certain maturity and are sufficiently risk averse such that they only trade within their desired maturity segment no matter how large of yield differentials exist at other maturities.

  1. Preferred Habitat theory

In this case, the investors are not that risk averse as in the M S T. In this case, they might depart from the preferred maturity if there is sufficient return to compensate them for the additional risk.

Example : (1 + y2 ) 2 = ( 1 + y1 )(1 + 1f1 + P )

P > 0

If there is relatively small demand for 2 year bonds, then 1f2 can be increased to boost that.

Thus there is a bias in the forward rate.

Preferred habitats :

Lenders borrowers

Short term Banks Corporations for inventory

Long term Insurance companies Corporations for long term assets

  1. Expectations theory

Investors’ expectation of future rates determines the shape of yield curve. Expectations of future increases in the short term rates leads to higher yields on long term bonds. In this theory, the forward rates are unbiased estimates of expected future spot future returns.

For a given time zone, the expected return from investment strategy of identical risk is identical If yield curve is upwards sloping investing in long term bonds does not mean a person will expect to earn higher returns. The difference in short term and long term bonds is not important to them.

  1. Liquidity preference theory

This theory predicts an upward bias in the yield curve.

It predicts that the liquidity premium is positive and increasing with maturity. In other words, observing an upward sloping curve does not mean spot rates will increase in the future.