We will discuss mathematics for options, pricing methods, and optimizing the performance of a portfolio. We will cover discrete (finite probability spaces, tree models) as well as continuous math (Brownian Motion, Black-Scholes formula, Martingale theory).
The theory in the course revolves around the “No Arbitrage theorem” which is closely related to one-price principle. Similarly, the idea of one-price principle theorizes that a set of investments with the same net outcome, albeit with different transactions, should have the same price.
Financial Markets
A financial market consists of tradable products such as stocks, bonds, currencies and indices. The market consists of two parties - the buyers and the sellers, who buy and sell these products respectively trying to make an arbitrage from the transactions. Formally, an arbitrage opportunity is a chance for a risk-free profit.
Odds and Arbitrage
Let us consider the example of a horse race, where P(S wins)=3/4 and P(W wins)=1/4. How do we calculate the odds against W winning? It is given by P(W wins)/P(W loses)=3. This situation is called 3-to-1 odds. That means, a dollar bet $1 brings $3 if W wins. Similarly, for S, it brings $0.3. These bets are an example of a fair game, the net reward 41(3)+43(−1) is 0.
A bookmaker sets the odds for such scenarios. Consider an other example where the odds are 9-to-5 against W and 2-to-5 against S. We claim that this results in an arbitrage for the bookmaker. To illustrate this, consider the bets $10 on S and $5 on W. The payoff for the bookmaker when S wins is 5−1052=1. When W wins, the profit is 10−559=1. A bookmaker typically decides the odds after the bets are made so that arbitrage is made in any scenarios.
Ideally, one wants to make profit from such games. How do we do this? In the first example, we considered the probabilities and calculated the odds. To decide the odds for arbitrage, we do the reverse. In the above scenario, given the odds, the probabilities are P(S)≈71% and P(W)≈36%. The probabilities add up to slightly above 100%, which makes up the arbitrage for the bookmaker.
Contingent Claim
A derivative or a contingent claim is a security whose value depends on the value of an underlying asset. Forward contracts, futures contracts and options are examples of such securities. Interestingly, the markets around the world also use derivatives whose underlying asset is also a derivative. Such securities are called as structured products.
With such contrived products, it is important to choose a fair price to ensure there are no arbitrage opportunities in the market. To do so, we introduce the required mathematical notation -
X is a real-valued random variable defined on a probability space (Ω,F,P) and G is a sub-σ-algebra of F. For unfamiliar readers, in probability theory, a probability space consists of three elements -
A sample space Ω, which is the set of all possible outcomes. For example, the sample space for a dice roll is Ω={1,2,3,4,5,6}.
An event space or a set of events is represented by F. For example, the events constituting an odd dice roll is F={1,3,5}.
A probability function P, assigns a probability (a number between 0 and 1) to each event in the event space.
The term σ-algebra on a set X refers to a nonempty collection Σ of subsets of X closed under complement, countable unions, and countable intersections. The pair (X,Σ) is called a measurable space.
In general, we model the price of a contract with a random variable X that evolves across time with X(w)≥0 (all the values for the random variable are positive). For example, for the call option described in #options, we have
X(w)=(S(w)t−K)+
Similarly, for a put option, where the holder has the right to sell a certain asset at a fixed price, we get
X(w)=(K−ST)+
There can be other kinds of structured products like
X=max(S1,…,ST)
X=(T1∑t=1TSt−K)+
In the above definitions, the variables w,St capture the evolution of price across time, and are described in the later section. The key takeaway is that, we are trying to price a contingent claim X - find C0(X)
Forwards Contract
A forward constract is an agreement to buy or sell an asset at a certain future time for a certain price. Consider the following example,
A agrees at time t=0 to sell one share at time T for $F. We assume that A buys the share at t=0 at price S0.
B agrees at time t=0 to buy one share at time T for $F
We assume that A,B don’t invest their own money, and borrow money from the bank to perform transactions. Given such a scenario, how do we decide a fair priceF? To simplify the calculations, we assume the following -
No transaction costs or dividends.
Market is liquid - every transaction has a buyer and seller available.
Investor is small compared to the market. That is, the action of the investor is not momentous enough to change the price of the stock in the market.
Short selling/borrowing of stock is allowed.
B can invest the amount in the bank for a continuous compound interest rate r during this term. The claim is that F=S0erT where S0 is the price of the share at t=0. To analyze this, we consider the following cases -
F>S0erT
A sells the share for $F and repays the loan of $S0 with interest - $S0erT. The profit A gets is F−S0erT
F>S0erT
B short sells one share for $S0 and invests $S0 in the bank. The money in the bank will grow to S0erT at t=T, and B buys one share at F. The profit for B here is S0erT−F>0
So the fair price is S0erT where neither A nor B can take advantage.
Note. A futures contract is similar to a forward contract but is traded on a financial exchange. They typically have a delivery month rather than a delivery date, and are followed by a settlement procedure called marking to market.
Options
An option is a contract which gives the holder of the option the right, but not the obligation, to buy other sell a given security at a given price (called the strike price) within a fixed time period [0,T]. A call option gives the option holder the right to buy at the given price, whereas the put option gives the option holder the right to sell at the given price.
European Call Options
A European option can only be exercised by the holder of the option at the expiration time T (unlike an American Option, which will be discussed in the later sections). We have the following equation for profit of a call option holder, assuming the option is free -
P=(ST−K)+={ST−K0ST>KST≤K
How do we decide the fair price at t=0 for the option? We need to use probability results to model the price movement to decide this. In discrete time settings, we use different models like Cox, Ross and Rubenstein. In continuous time frame, we use Brownian motion using the Black-Scholes model. In both cases, there exists a unique fair price. However, if the product is slightly more complicated, then there is no unique fair price.
Binomial Model
Also known as Cox-Ross-Rubenstein Model (CRR) is a discrete time model for a financial market defined for t=0,1,…,T. At each time instant the model considers the following two types of assets
Bonds
Considered as a risk-free asset, a bond yields at a constant rate of return r≥0 over each time period. The price is modeled as
Bt=B0(1+r)t
for t=0,1,…,T. B0 is typically assumed as 1.
Stocks
Stocks are considered as risk-models where the price evolution is captured using binomial random variables. We consider the model
St=S0ζ1…ζt=St−1ζt
where S0>0 is the constant initial price. That is, the stock price process is modeled as an exponential random walk.
In this model, we assume P(ζt=u)=p and P(ζt=d)=1−p where 0<d<1+r<u where d,u are the down and up movements respectively. That is, St+1=Stu represents the price of the stock if the price has moved upward at time t.
Concretely, the probability space (Ω,F,P) is such that Ω is the finite set of 2T possible outcomes for the values of stock price (T+1)-tuple (for S0,S1,…,ST), F is the σ-algebra consisting of all possible subsets of Ω, and P is the probability measure on (Ω,F) associated with the Bernoulli probability p.
This is a partition of Ω. Similarly, for t=2, we get
F1={ϕ,Ω,{uu,ud},{dd,du}}. Extrapolating this,
Ft=σ(S1,…,St)=σ(ζ2,…,ζt)
We get #(F)t=2t. This evolution can be represented using a binomial tree of the following form -
dddduuuduu…
Trading Strategies and Fair Pricing
A strategy ϕ in stocks and bonds consists of sequence of random variable pairs ϕ={(αt,βt);t=1,…,T} where αt represents the holdings of a stock and βt represents the holdings of a bond on day t (The unit for time can be taken as minutes, hours, etc., but we consider days). We also have the following assumptions
Predictability: αt,βt∈Ft−1,t=1,…,T
Self-financing: αtSt+βtBt=αt+1St+βt+1Bt. That is, there is no influx or outflux of money asides from the changes in the underlying assets. We have some initial investment and assume that α1,β1 are F0-measurable are constant.
The value of the portfolio at time t is given by
Vt(ϕ)=αtSt+βtBt,t=1,…,T
and
V0(ϕ)=α1S0+βB0=α1S1+β1
We can now define the concept of arbitrage formally.
Arbitrage
An arbitrage is a trading strategy ϕ such that V0(ϕ)=0 and Vt(ϕ)≥0. That is, we don't have any initial investment and get a profit at time t=T. However, note that Vt(ϕ) is a random variable, so the inequality is assumed to be a point-wise inequality wherein VT(ϕ)(w)≥0,∀w∈Ω. This is a stronger assumption than P(VT(ϕ)>0)>0≡E(VT(ϕ))>0≡VT(ϕ)(w)>0 for some w∈Ω.
Notice that the last equivalence does not have any probability term in the expression. This equivalence is important to note since it makes the definition invariant of the underlying probability assumptions.
Hedge
A hedging or a replicating strategy ϕ has the property VT(ϕ)=X. That is, to price a contingent claim X, if we find a trading strategy ϕ with cost (initial investment) V0(ϕ), such that VT(ϕ)=X, then V0(ϕ) is a fair price of X by the one-price principle. Essentially, with the price “V0(ϕ)” we bought an opportunity that resulted in a value of "VT(ϕ)" at time T. If a contingent claim X has the same value at time T, then its fair price should be V0(ϕ) at t=0.
Using this as a motivation, let us price a European call option. Our approach is as follows - an investment in the underlying stock and a bond should result in the same outcome as an investment in the option itself. The movements in the option will be correlated with those of the underlying stock, allowing us to correctly determine a fair price at every time instant.
Hedging for T=1
Consider the following example, where we try to hedge X at t=1. That is, we are trying to find a strategy ϕ=(α1,β1) such that V1(ϕ)=α1S1+β1B1≡X.
We have two cases based on the stock movement -
α1S0u+β1(1+r)=Xu
α1S0d+β1(1+r)=Xd
Equivalently, α1S0(u−d)=Xu−Xd
α1β1=u−dXu−XdS01=u−duXd−dXuB11
What do the above expressions mean? Investing α1 in stock and β1 in bond at the end of day 0 will result in same price as X on day 1. Manipulating the expressions further,
The above expression is reminiscent of an expectation formulation where p∗=u−d1+r−d and 0<p∗<1. Here, the set of discounted stock prices {S0,S1/(1+r)} form a p∗-martingale.
What is a Martingale? A Martingale is a sequence of random variables representing a stochastic process for which, at a particular time, the conditional expectation of the next value in the sequence is equal to the present value, regardless of all prior values.
It is important to realize that computing expectations under p∗ is a mathematical device. We are not assuming that the stock price actually moves according to this probability. That is, p∗ may be unrelated to the subjective probability p that we associate with the binomial model for movements in the stock price.
Theorem 1 (T = 1)For a given X, the value E∗(X)/(1+r) is the non-arbitrage price of X at t=0.Proof.
Let C0 is the market price of the contingent claim X and V0=E∗(X)/(1+r)=V0(ϕˉ) where ϕˉ=(αˉ1,βˉ1) (the values we calculated before). Consider the following cases -
Say C0>V0
To make an arbitrage in this situation, we sell X for C0. We invest the amount V0 in the stock and amount C0−V0>0 in bonds at t=0. At time t=1, we get V1(ϕˉ)=X yielding (C0−V0)(1+r)>0 as profit from the bond.
Say C0<V0
We do the opposite from above, where we buy the asset X. Short sell ϕˉ for V0, and invest C0 in X and V0−C0 in the bond. We get the profit (V0−C0)(1+r) at t=1.
Say C0=V0
Here, we can show that no one can make an arbitrage. Consider the strategy in stock, bond and contingent claim ψ=(α1,β1,γ1). We need to argue that no arbitrage can be made with any such strategy. We have,
We can distribute the expectation this way due to the linearity property. Recall that, we typically assume B0=1 for brevity. Continuting the calculations,
If V0(ψ)=0, then E∗(V1(ψ))=0, and if V1(ψ)≥0, then P∗(V1(ψ)=0)=1. That is, V1(ψ)=0 meaning no arbitrage can be made with any strategy ψ.
General T
This analysis can be extended to derive the hedging strategy a general t>1. Given an X, find ϕ={(αt,βt):t=1,…,T} such that VT(ϕ)=X (resulting in C0=V0(ϕ)). To derive such a strategy, we work backwards in time starting from VT(ϕ)=X. This way, we can also calculate the intermediate price - the price to enter the market at some intermediary time t. It is helpful to think of the tree visualization for the price evolution.
VT(ϕ)=αTST+βTBT=X
conditional on FT−1. Focusing on T−1 and T, the analysis is similar to the previous case wherein XU=X(w1…wT−1u) and Xd=X(w1…wT−1d). Doing so we get
The above relation can also be alternately derived using the Tower law to relate the value with Vt(ϕ) for a general t.
Tower Law: If X is a random variable whose expected value E(X) is defined, and Y is any random variable on the same probability space, then E(X)=E(E(X∣Y)), i.e., the expected value of the conditional expected value of X given Y is the same as the expected value of X.
Given (Ω,D,P) with g⊂F, X∈L′(F) (the space of random variables that are integrable with respect to the σ-algebra F) then Y=E(X∣g) is the random variable such that
Y∈L′(g)
E(YZ)=E(XZ) for all Z that are g-measurable.
where L′ represents integrability. The above property is the general statement for the law of general statician.
Special case. When Ω is finite (#(Ω)<∞), then F={all subsets of Ω} and G=σ(G1,…,Gn) form a partition (Gi∩Gj=ϕ,∪j=inGj=Ω) of Ω with P(Gk)>0.
A random variable Z is g-measurable iff Z(w)=∑k=1n1Gk(w)⋅ck≡Z(w)=ck if w∈Gk where ck is the cardinality of Gk.
Then,
E[X∣G](w)=k=1∑n(1Gk(w)⋅E[X∣Gk])
From Bayes' rule
E(X∣Gk)=E(X.1Gk)/P(Gk)=E(X;Gk)/P(Gk)
Using the above results, we get
X→E(X∣Y) is linear & positive (X≥0⟹E(X∣g)≥0)
E(E(X∣g))=E(X)
E(XZ∣g)≡E(X∣y)Z
If E∈g, then E(X∣g)=X and X⊥⊥g, then E(X∣Y)=E(X).
Exercise.
Show that E(⋅∣G) is a ⊥ projection where d(X,X~)=E(X−X~)2
Tower Property
For any G⊂H(⊂F), we have E[E[X∣H]∣G]=E[X∣G].
What does G⊂H mean? In the discrete case where G=σ(G1,…,Gn) and H=σ(H1,…,Hm), we have Gk=∪˙j∈J(k)Hj where J(1),…,J(n) is a partition of {1,…,m}.
In the trading strategy context, we have F0⊂F1⊂⋯⊂Fn.
Martingales
We now introduce Martingales formally - A Martingale is a sequence of random variables {M0,…,MT} such that
Adapted:Mt∈Ft, for all t
Integrable:E(∣Mt∣)<∞ for all t
E(Mt∣Ft−1)=Mt−1, 1≤t≤T. From Tower law, E[MT∣Fs]=Ms.
Let us now revert to our discussion on trading strategies with these formal definitions in mind.
As we have seen before in #General T, ϕ that satisfied Vt∗(ϕ):=Vt(ϕ)/(1+r)t is a p∗-martingale, where p∗=u−d1+r−d. A special case of this is St∗ is a p∗- martingale. We have the following theorem,
Theorem 2. (General T)V0 is the arbitrage-free price of V
Proof. We follow a similar structure as before,
C0=V0, then arbitrage is available
C0=V0, we have for any strategy ψ={(αt,βT,γ1):t=1,…,T}, V0(ψ)=0,VT(ψ)≥0 due to self-financing. Let ϕˉ={(αˉt,βˉT):t=1,…,T} be the hedge for X. Then, VT(ϕˉ)=X and V0(ϕˉ)=V0.
Let us consider a numerical example to understand this better - Consider the call option X=(ST−K)+, where T=4,u=1.2,d=0.8,r=0 (p∗=0.5). Let S0=50 and K=40. What is the fair price of X?
We construct the tree recursively, starting from the bottom-most level. At time T, we know VT, and then we use Vtup∗+Vtd(1−p∗) to get Vt−1(1+r).
Call-Put Parity
A call option is represented as CT=(ST−K)+ and a put option is represented as PT=(K−ST)+=(ST−K)−.
For any b, we have b+−b−=b and b++b−=b. In other words, the payoffs in call and put options are related to each other as
CT−PT=ST−KCT+K=ST+PT
Claim. For t<T, we have
Ct+(1+r)T−tK=St+Pt
This result can be proved using One price principle. Think of each side of the equation as separate investments
First investment has a call option and bonds with K.
Second investment has a put option and one share of the stock
We can see that at time T, the values become equal CT+K=ST+PT
American Options
In American Options, we have Y0,Y1,…,YT random payouts where each Yt∈FT. The holder of the options can take the payment at any time t∈[0,T] of choice. That is , unlike European options, American options have the choice of payment at any intermediate time. For example, consider a call option Yt=(St−K)+ for t=0,…,T.
Formally, we have the stopping time property represented as T:Ω→{0,1,…,T} such that {T≤t}={w∈Ω:T(W)≤t}∈Ft,∀t=0,1,…,T. This is equivalent to saying, {T=t}∈Ft,∀t.
For example, consider Z0,Z1,…,ZT, we have T:=min{t:ZT∈B}∧T where A∧b=min(a,b). Then, {T=t}={Z0)∈B,Z1∈B,…,Zt−1∈B,Zt∈B}∈Ft.
Optional Stopping Theorem
Consider a stopping time τ and a Martingale (Mt), we have E(Mτ)=E(M0). Essentially, in a fair game, the average payout from exiting the option at an intermediate time is the same as the average payout till time t=T.
Here Mτ is a random variable of the form Mτ(w)(w) which we have represented in a singular form.
Is the converse true? That is, a sequence of random variables have the optional stopping property where expectations are the same for the beginning and an intermediate time. Then do they constitute a Martingale? Yes, we have
Let us consider an example of an American Put option to understand this better -
Yt=(K−St)+
where T=1,u=2,d=0.5,r=0.2,S0=100,K=150. The value for p∗ is 7/15. Then, we have the following tree evolution for St
5025100100100100400
The payment Yt looks like
100112550500500
How do we calculate a fair price in this case? We introduce the notion of a super-martingale. In comparison, in European call options, we get for Vt
7511255041.9822.2ˉ500
From the seller's perspective, we have Ut, and we use something called as a necessary wealth process - calculating the amount of money the seller requires at every node to pay the option holder. To do so, we need to consider the maximum of all possible payouts at every node. In the node labeled 1 above, if the buyer exits the option at t=1, then the seller needs a payout of 100. However, if the buyer exits at time t=2, the seller requires atleast 75 at t=1 (They can invest the 75 in the same stock using short sell to get the required payouts). Considering these scenarios, the seller requires 100 at the 1 node. Doing such a process for every node, we get Ut
1001255053.0922.2ˉ500
We can see that the price of this option is slightly higher than European since the buyer has the choice of exiting the contract at any intermediate time.
To formally describe the process we did above, we consider the following recursion
How do we now show that this is minimal? Zt being minimal implies that if Zt is a supermarginale majorant of {Yt}, then Xt≥Zt for all t.
Xt(w)≥Yt(w)∀t,∀w⟹Xτ(w)(w)≥Yτ(w)(w),∀w
Then, equivalently, Xτ≥Yτ for t≤τ≤T.
Xt≥E(Xt∣Ft)≥E[YT∣Ft]≥E[YT∣Ft],∀t
Resulting in Xt≥Zt. So far, we have shown (1) and (2) in the claim. Also, in the above proof, we use the following property. Let Xt be a supermartingale ifor t≤τ≤T. Then E(Xτ∣Ft)≤Xt. The proof is left as an exercise.
How do we show (3) and (4) in the claim. For (4), we simply need to check the martingale property
An important to thing to note here is that these values are random variables and are measurable under Ft−1. However, this strategy will not be self-financed due to the super-martingale property. The value of the portfolio is
This trading strategy ϕ∗={(α∗,β∗):t=1,…,T} is a super-hedging strategy with Vt(ϕ∗)≥Ut. We introduced δ~t to make it a self-financing strategy.
Define τˉ=τˉ(0)=min(T:Ut=YT)∧T. Then, Ut−1>Yt−1 and δ~t=0 for t=1,…,τˉ(w). Consequently, Vt∧τ(ϕ∗)=Ut∧τ, and Ut∧τ∗=(1+r)t∧τUt∧τ forming a martingale.
Arbitrage
For a trading strategy ψ, we have V0(ψ)=0 and VT(ψ)≥0. Alternately, P(VT(ψ)>0)>0. In an American option, there is an asymmetry between the buyers and the sellers. The buyer has to choose only one time t≤T to exit the option whereas the seller has to safeguard against an arbitrage in all possibilities.
The seller's arbitrage for a strategy ϕs (in stocks and bonds) is then defined as
V0(ϕs)∀τVτ(ϕs)=C0≥Yτ>sometimes
The seller sells an option for C0 and makes an investment of C0 in ϕs. Then an arbitrage for the seller is
VT=(Vτ(ϕs)−Yτ)(1+r)T−τ
where τ is the time when the buyer exits the option. This value is always greater than or equal to 0 with a non-zero probability that it is strictly greater than 0.
On the other hand, the buyer's arbitrage for a strategy ϕb is defined as
V0(ϕb)∃τVτ(ϕb)+Yτ=−C0≥0>sometimes
The arbitrage at time T is (Vτ(ϕb)+Yτ)(1+r)T−τ.
What would be the good time to exercise the option? In the following sections, we'll see that τ=τˉ is the optimal time to exit the option.
Example
Consider a put option YT=(K−St)+ where K=150,S0=100,u=2,d=0.5,r=0.2,T=2. We get U0=53.09 and Y0=50. Then, the super-hedging strategy is given by
To summarise, for a payout random variable Y={Y0,…,YT}, we have Ut as the necessary wealth at each time t. Then, the super-hedging strategy ϕ∗ has the following properties -
Vt∗(ϕ∗) is a martingale
V∗(ϕ∗)≥Ut∗≥Yt∗ for all t
We define τˉ such that τˉ=min{t∣Ut=Yt}∧T. With this definition, we have
Vt∗(ϕ∗)=Ut∗
for t=0,1,…,τˉ and V0(ϕ∗)=U0.
Theorem.U0 is the no-arbitrage price for American call option based on Y.
Proof.
C0>U0, then there's arbitrage for the seller. The seller sells the claim for C0 and buys ϕ∗ for U0. The remaining amount, C0−U0 is invested in bonds. Then, for this strategy ϕs, we have
The relation between call and put options in general can be shown using Jensen's inequality. We have
E(ϕ(X))E(ϕ(X∣G))≥ϕ(E(X))≥ϕ(E(X∣G))
The payout functions for both call and put options are convex. We aim to show that Ut=Vt for t=T,T−1,…,1,0, using induction. The statement is true for t=T.
The binomial model we have discussed previously is an example of a finite market model. These structures can be generalised to markets with multiple stocks, bonds and contingent claims. The price evolution need not be binary We shall derive results for pricing similarly like before for this general market. To do the analysis, the notation for probability space, portfolio, value, trading strategies and arbitrage has to be defined -
Probability Space
Again, we consider a finite probability space (Ω,F,P), that is #(Ω)<∞) with F=P(Ω) (the power set) and P(A)=∑w∈AP(w),A∈F.
We shall assume that every event has a non-zero probability. If there are events with zero-probability, they are not relevant to our discussion and we can simply prune these points.
A filtrationF0⊂F1⊂⋯⊂FT represents the history of the market or the evolution of prices of assets in the market. Typically, FT=F.
Assets in portfolio
Now in this setup, we consider the assets - asset 0 bond and assets 1,2,…,d stocks. At any time t, the prices are given by
St=(St0,St1,…,STd),t=0,1,…,T
Where St0 is the bond-price and not random and StU for i>0 represent the stock prices.
Each sequence {Sti}t=0T is Ftmeasurable, that is (St) is adapted to the filtration.
Trading strategy
After defining the prices and the structure of the portfolio, we define a trading strategy for this market -
ϕ={ϕt:t=1,…,T}
where ϕt=(ϕt0,ϕt1,…ϕtd) is a vector of holdings representing the assets held at time t. That is, ϕtI is the number of shares of asset i to be held on day t.
The basic assumptions with our trading strategies are
The trading strategy is predictable - ϕti∈Ft−1 for all i and t=1,2,…,T.
Self-financing - ϕt⋅St=ϕt+1⋅St (note that these are vector dot products ϕt⋅St=∑i=1dϕtiSti).
Value of the portfolio
The value of the portfolio at time t=0 is given by V0(ϕ)=ϕ1⋅S0 and for time t>0, we have Vt(ϕ)=ϕt⋅St. For this market, the change in the value of portfolio across a day is given by ΔVt(ϕ)
To help us derive results in the later sections, we define the following notation for discounted stock prices -
St∗=St/St0=(1,St1∗,…,Std∗)
It essentially divides the stock prices with the bond price at time t. Similarly we can define discounted value of a portfolio.
We can derive that the telescopic sum property still holds with the discounted values for the portfolio. That is,
VT∗(ϕ)=V0∗(ϕ)+u=1∑tϕu⋅ΔSu∗
where ΔSu∗=Su∗−Su−1∗. This property will be useful for extracting Martingales from the price evolution tree. Specifically, we have the following lemma
Lemma 1 If (Mt:t=0,1,…,T) is a martingale and (H1,H2,…,HT) is a predictable sequence of random variables then
Xt:=u=1∑THuΔMu
is a martingale!
Arbitrage
The expressions are very similar to before wherein an arbitrage opportunity (risk-free profit) is a trading strategy ϕ such that
V0(ϕ)VT(ϕ)P(VT(ϕ)>0)=0≥0>0
The last condition essentially ensures that there is atleast one case where the profit is non-zero. Previously, we had used expectation to ensure this using the p∗ model derived using the binomial model. However, we want a general scenario here and an expectation requires an associated probability distribution.
The market is set to be viable if there are no arbitrage opportunities. We can show that the binomial model considered previously is viable if d<1+r<u. So, when is a market non-viable? Consider Ω={w1,w2},T=1,d=2 with the stock price evolution S00=S10=1, S01=S02=1, and
( w_1 )
( w_2 )
( S_1 )
1
2
( S_2 )
1
3
The trading strategy ϕ={(0,−1,1)} has V1(ϕ)≥0 in all entries and >0 in at least one entry.
Notice the subtleties in this example. Previously, in the CRR model, we focused on finding an arbitrage-free pricing. We ensured there is no arbitrage in the market with this design. However, considering a trading strategy with only stocks in a bullish market, there is always an opportunity for arbitrage.
For trading strategies with only stocks, the problem of finding a trading strategy boils down to solving some linear inequalities.
Martingale Measure
As noted earlier, we will assume that P(w)>0 or each w∈Ω. For a new measure (Ω,F,Q), we say that Q is equivalent to P (Q∼P) if for all A∈F
Q(A)>0⟺P(A)>0
With this infrastructure, we are aiming to find a new probability Q (like p∗ in CRR model) to prove properties in a market.
Definition. A probability Q on (Ω,F) is a martingale measure provided {STi∗:t=0,1,…,T} is a Q−martingale for each stock I=1,…,d.
An equivalent martingale measure is a probability Q such that Q∼P and Q is a martingale measure.
With these tools, we now define the first fundamental theorem of asset pricing.
First Fundamental Theorem of Asset Pricing
A market is viable if and only if there exists at least one Equivalent Martingale Measure (EMM).
To understand this better, consider the previous example of a market that is not viable. Suppose we have a martingale measure Q, then
EQ(Si1)=S0i
for I=1,2 using the martingale property. Letting a=Q(w1),b=Q(w2), we get the linear equations
a+2ba+3b=1=1}b=0
Since P(w2)>0⇒Q(w2)>0, there cannot exist an Equivalent Martingale Measure in this market.
Proof
(⟸) Let Q be an EMM and ϕ is a strategy making an arbitrage. That is,
V0(ϕ)=0,VT(ϕ)≥0
Since {St}t=1T is a martingale under Q, the sequence ∑i=1tϕuI⋅ΔSui∗ for t=0,1,…,T is a Q-martingale.
Vt∗ is also a martingale under Q, implying EQ(VT∗(ϕ))=EQ(V0∗(ϕ))=0. Since P∼Q, P(VT(ϕ)>0)=0.
(⟹) Now, we show the opposite direction, that is for a viable market, there exists an EMM.
We have #(Ω)<∞, where Ω={w1,w2,…,wn}. Consider a random variable X:Ω→R, then we have (X(w1),X(w2),…,X(wn)∈Rn) and a distribution Q with probabilities (Q(w1),…,Q(wn)). Then, we have
EQ(VT∗(ϕ))=k=1∑nQ(wk)VT∗(ϕ)(wk)=Q⋅VT∗(ϕ)
It suffices to show that there exists a Q such that Q⊥{VT∗(ϕ):ϕ is a trading strategy with V0(ϕ)=0}⊂Rn}.
We have seen before that as a result of a telscopric sum formulation, VT∗(ϕ)=V0∗(ϕ)+∑u=1Tϕu⋅ΔSu∗.
Lemma 2
Suppose a M=(M1,…,Mn) is an adapted sequence such that
E(u=1∑Tqu⋅ΔMu)=0;ΔMk=Mk−Mk−1
for all predictable (q1,…,qT), then M is a martingale.
Proof. To show, E(Mt∣Ft−1)=Mt−1 for t=1,2,…,T. Equivalently, E(Mt⋅1A)=E(Mt−1⋅1A) for all A∈Ft−1.
If X,Y are both G-measurable and E(X⋅1A)=E(Y⋅1A) for all A∈G, then P(X=Y)=1 (Consider {X<Y}=A).
In our context X=E(Mt∣Ft−1) and Y=Mt−1, leading to (ΔMt⋅1A)=0. Fixing t and A∈Ft−1, let
qu={1A0u=tu=t
Consequently,
u=1∑Tqu⋅ΔMu=1A⋅(Mt−Mt−1)
Lemma 3 If we have, qti∈Ft−1 for i=1,…,d amd t=1,…,T, then for any c∈R, there exists ϕt0 for t=1,…,T such that
ϕ={(ϕt0,ϕt1,…,ϕtd):t=1,…,T}
is a self-financing trading strategy and V0(ϕ)=c.
Proposition. Suppose Q∼P, such that EQ(VT∗(ϕ))=0 for all trading strategies ϕ with V0(ϕ)=0, then Q is an EMM. That is, we need to show St1∗ is a Q-martingale. From lemma 2, it is equivalent to showing EQ(∑u=1Tqu⋅ΔSu1∗)=0 for all (q1,…,qT). Consider the following strategy
Now, for all trading strategies with the self-financing property, the set L:=VT∗(ϕ):ϕ with V0(ϕ)=0⊂Rn is a linear subspace of Rn. Similarly, consider the set D:={Y∈Rn:Y≥0 with Y=0}.
Our goal boils down to show that a viable market exists if and only if L∩D=ϕ. It is equivalent to show L∩F=ϕ for F:={Y∈D:∑i=1nYk=1}.
Assume viability -
We have noted that L is a subspace of Rn. Also, F is compact and convex.
Lemma.Separating Hyperplane Theorem For a vector subspace L⊆Rn and a compact convex set F, if L∩F=ϕ, then there exists a ζ∈Rn, ζ=0 such that
LF⊂{t:t⋅ζ=0}⊂{y:y⋅z>0}
Proof. Let G:=F−L={f−l:f∈F,l∈L}, then G is convex and closed and 0∈G. We choose the vector ζ∈G that minimizes ∥x∥, x∈G. Now,
β:=inf{∥x∥:x∈G}
then for any sequence xn∈G such that ∥xn∥ converges to β (∥xn∥↘β) , using the parallelogram law
∥xn−xm∥2=2(∥xn∥2+∥xm∥2)−42xn+xm2
The terms ∥xn∥2,∥xm∥2 converge to β2 and the average 2xn+xm2 goes to β2 leading to ∥xn−xm∥2→0 for m,n→∞ and xn→ζ. Therefore, {xn} form a Cauchy sequence convergent to some ζ. Since G is closed, this shows the existence of such ζ∈G.
Now, for any x∈G, x⋅ζ≥∥ζ∥2. How do we show this? Consider any arbitrary vector formed by x,ζ, ∥αx+(1−α)ζ∥2−∥ζ∥2≥0 (a quadratic in α, and 0≤α≤1). For α→0, we get
−2∥ζ2∥+2(x⋅ζ)≥0
Now, in particular, since 0∈L, if f∈G, then f∈F. Consequently, if f∈F, then for any λ∈R,l∈L, f−λl∈G. We get
f⋅ζ≥λ(l⋅ζ)+∥ζ∥2
With the limits of λ, we can conclude that l⋅ζ=0. □
How does this theorem help us show L∩F=ϕ for a viable market? Consider the EMM Q(wk)=cζk>0. c=∑i=1nζi. For a fixed k, consider
Yi={10i=ki=k
Then for Y∈F, Y⋅ζ>0. Let ϕ be a trading strategy with V0(ϕ)=0, then
EQ(Vt∗(ϕ))=Q⋅VT∗(ϕ)=c1ζ⋅∈LVt∗(ϕ)=0
Completeness of a Market
A merket is complete if for every contingent claim X∈FT, there exists a trading strategy ϕ that can be replicate the claim with VT(ϕ)≡X.
Second Fundamental Theorem of Asset Pricing
Assuming the market is viable, then the market is complete if and only if there is a unique EMM for the market.
Proof.
(⟹) Assume that the market is complete. Let P1∗,P2∗ be the distinct EMMs. Then for a A∈FT, define X:=1A. Due to completeness, ∃ϕ with VT(ϕ)=1A.
Pi∗(A)=Ei∗(1A)=E1∗(VT(ϕ))=S00ST0⋅V0(ϕ)
Therefore, P1∗(A)=P2∗(A) for all A∈FT implying that P1∗=P2∗.
(⟸ ) Assuming that there is a unique EMM, we need to show that the market is complete. We do this by proving the contrapositive - a market is incomplete implies that there is no unique EMM for the market. We define
L:={VT∗(ϕ):ϕ self-financing trading strategies }
Now, we show that the market is incomplete if and only if L⊊Rn and ∃z=0 such that z⊥L. Consider the EMM Q for a viable market. Then, define Q~(wk)=Q(wk)+czk for some c>0. Then,
k=1∑nQ~(wk)=k=1∑nQ(wk)+c(z⋅1)=1+0
Is 1∈L? Yes, the strategy of investing in one bond yields a discounted value of 1 at T.
Now, it is left to show that there exists a c such that Q~(wk)>0. Consider the following formulation
This proves the second fundamental theorem of asset pricing.
Note that if Q1,Q2 are two EMMs, then a linear combination of these αQ1(A)+(1−α)Q2(A) for A∈F and 0<α<1 is also an EMM.
Example
For T=1,Ω={w1,w2,w3}, let S10=S00=1, S01=2 and S11=(135)T. To check if the market is viable, we try and solve for an EMM on this market - Let Q=(a,b,c), then
a+b+c=1,a>0,b>0,c>0a+3b+5c=2
Solving, we get
Q(c)=(21+c,21−2c,c)
for 0<c<41. This shows that the market is viable. Also, the market is not viable since there are multiple EMMs possible.
That means, when x1+x3−2x2 is 0, then the expectation is independent of c - X can be replicated in some cases even if the market is incomplete!
Let us try an replicate such an X. Consider the bond holdings α and stock holdings β.
αS11+βS10α135+β111=X=x1x2x3
X is replicable iff X∈span((135),(111))≡X⊥(1−21). This again yields x1+x3−2x2.
Theorem. In a viable and complete market, the number V_0(\phi) is the no-arbitrage price of a European Call Clain X = V_T(\phi).
Note. We have seen this before in the context of a binomial model, but this theorem is for a general market with discrete possibilities. In our previous proof, we did not use any binomial properties but rather showed the result using martingale properties. The same proof follows for this as well.
How do we conclude such results in a general case? Even when the market is incomplete, how do we find contingent claims which can be replicated?
Consider the set M={Q:Q is an EMM}=ϕ. Then for any measure in this set,
Suppose X is replicable, then X=VT(ϕ~) for some ϕ~. Then,
V0(ϕ~)≥V+(X),V0(ϕ~)≤V−(X)V−(X)≤V+(X)
leading to V+(X)=V−(X)!
Does the converse hold? In the previous example, say we have a contingent claim X=(1925)=(S11)2. Now consider a strategy ϕ=(α,β), we have
V0(ϕ)V11(ϕ)α135+β111=2α+β≤X for V−(X)≤1925
Solving these inequalities, we get V−(X)=5, and similarly, V+(X)=7. Let us check if X can be replicated
E(c)(X)=5+8c,0<c<41
It is not possible for the expectation to be independent of c. Also note that the range of the above expression is exactly (5,7).
We have seen that V+(X)≥V−(X) for a contingent claim X.
If C0>V+(X) then there is a seller's arbitrage. How? There exists a strategy ϕ such that VT(ϕ)≥X and V0(ϕ)=V+(X)<C0.
Similarly. when C0<V−(X) there is a buyer's arbitrage.
Theorem. A contingent claim X is replicable if and only ifV+(X)=V−(X)
Proof.
(⟹) We saw that previously
(⟸) Assume V+(X)=V−(X). Let ϕ+ be a trading strategy such that VT(ϕ+)≥X and ϕ− is such that VT(ϕ−)≥X. However, V0(ϕ+)=V+(X)=V−(X)=V0(ϕ−) and consequently VT(ϕ+)−VT(ϕ−)≥0. Let Q∈M be any martingale measure -
Therefore, VT(ϕ+)≡VT(ϕ−)≡X implying that ϕ+ can be used to replicate X.
Suppose X is not replicable, then there is ϕ such that VT(ϕ)≤X and V0(ϕ)=V+(X).
Claim.P(VT(ϕ)<X)>0. If it is equal to 0, then VT(ϕ)≡X.
So, any market price for X with C0∈(V−(X),V+(X)) yields an arbitrage.
Theorem.X is replicable if and only ifEQ(X) is the same for all Q∈M.
Proof.
(⟹)X=VT(ϕ) then
EQ(X)=EQ(VT(ϕ))=V0(ϕ)S00ST0
(⟸) Assume for contradition X is not replicable. As we've constructed before, let L={VT(ϕ):ϕ is a trading strategy}. Since X is not replicable, X∈L. Let X=X0+Z such that X0∈L and Z=0,⊥L. For any Q∈M, we define Q~(wk)=Q(wk)+cZk. Again, we choose a c such that Q~>0 and Q~∈M. Then, since EQ(X) is the same for all Q∈M, we have
EQ(X)Q⋅X=EQ~(X)=Q~⋅X=Q⋅X+cZ⋅X∴Z⋅X=0
implying that X∈L giving a contradiction.
Dual Formulae
As a result of the above theorem, we can give alternate formulae to V+(X) and V−(X) (dual comes from linear programming)
V+(X)V−(X)=sup{EQ(X):Q∈M}=inf{EQ(X):Q∈M}
Random Walk
In this section, we shall extend our theory beyond finite market models, wherein the price evolution is described using a random walk.
Let Xn=ξ1+⋯+ξn where E(ξk)=μ and ξ are iid for n=0,1,2,…. Let Fn=σ(ξ1,…,ξn) and Fn⊥⊥ξ+n+k for k≥1. Then, consider set of random variables Mn=Xn−nμ.
Brownian motion is defined as a stochastic process defined on continuous time - W=(Wt,t≥0). It is defined on a probability space (Ω,F,P) and has the following properties -
The initial value is assumed to be W0=0
Independent increments - Given a set of times t0,…,tn such that 0=t0<t1<⋯<tn then Wt1−Wt0,…,Wtn−Wtn−1 are independent.
Gaussian increments - For 0≤s≤t, Wt−Ws∼N(0,t−s).
For a fixed ω∈Ω, the function t→Wt(ω) is continuous.
Equivalently, a Brownian motion is a stochastic process (wt:t≥0) is a Gaussian process with E(Wt)=0,cov(Ws,Wt)=s∧t for all s,t and t→Wt is continuous.
This is equivalent to the third property, for s<t,
Brownian motion is a unique stochastic process that has some symmetry properties. That is, transformations that yield a Brownian motion when applied on a Brownian motion. Let (Wt) be a Brownian motion, then the so are the following
Scaling - c1Wct for c>0
Reflection - −Wt
Time Inversion - tW1/t with limt→0tW1/t=0
Time Shift - W~t=Wt+s−Ws, t≥0 for a fixed s. This comes in handy to define Markov processes.
Time Reversal - For a T>0, W~t=W~T−t−WT for 0≤t≤T.
More Properties
Define the sigma field Fs:=σ(Wu:0≤u≤s), then (Wt+s−Ws)t≥0⊥⊥Fs - This is the Markov Property for Brownian Motion.
Brownian paths are not smooth
Wt+h−Wt=O(h)
roughly whereas if f is a C′ function,
then f(t+h)−f(t)=O(h). h≫h as h→0. A Brownian motion path has a property called fractal and it is nowhere differentiable.
Also,
k=1∑2n[W(k2−nt)−W((k−1)2−nt]2→t
as n→∞. For a smooth function f, doing the same thing yields 2−nt→0. This observation forms one of the fundamental building blocks for stochastic calculus.
Another peculiar property of Brownian motion has is -
tWtt→∞limsupWtt→∞liminfWt→0=∞=−∞
Since Brownian motion is basically the continuous version of random walks, it has some Martingale properties. Specifically, the following processes are martingales
WtWt2−texp(zWt−z2t/2) for z∈C
The last result comes from the properties of moment generating functions - for a random variable X∼N(0,1), we get E(ezX)=ez2/2. Then,
Wiener also studied the integration properties for Brownian motion
Mtf=∫0tf(u)dWu=Wtf(t)−∫0tWuf′(u)du
for a f∈C′ and non-random. The above expression comes from integration by parts and is called as the Wiener's integral. Also, Mtf is a martingale (This is easy to show).
Doob's inequality - Let (Mt) be a martingale with right-continuous paths. Then, for each t>0,
E[0≤s≤tsupMs2]≤4E[Mt2]
Passage times - Let τb=min(t:Wt=b), then
Ex(e−ατb)=exp(−2α∣b−x∣)
where Ex means we use x+Wt (we are changing the initial mean value of Wt).
Proof. Fix λ>0 and x<b. Since exp(λWt∧τb−λ2(t∧τb)/2) is a martingale, and is ≤eλb, we get
Taking λ=2α, we get the statement of the theorem. Observe that when α→0, the RHS becomes 1. Therefore, Px(τb<∞)=1 as α→0. That is, Brownian motion will eventually get to any level!
Wiener's integral - Formally,
Mtf:=f(t)Wt−∫0twsf′(s)ds
It is useful to know the distribution of this process. Thinking of the above integral as Riemann's sum, we can see that Mtf has the distribution N(0,σ2). The variance is calculated as
E[(Mtf)2](2)(2)+(3)E[(Mtf)2]=f(t)2t+(2)∫0t∫0t(u∧v)f′(u)f′(v)dudv−(3)2f(t)∫0tuf′(u)du⟹2∫0t(∫0uf′(v)dv)uf′(u)du∵ by symmetry=2∫0t[f(t)−f(u)]uf′(u)du⟹−2∫0tuf(u)f′(u)du=−tf(t)2+∫0tf(u)2du=∫0t[f(u)]2du
This is a very useful result. In fact, using two different functions -
MtfMtg−∫0tf(u)g(u)du
is also a martingale! An easier way to show this would be using polarization - manipulate (Mtf+Mtg)2−(Mtf−Mtg)2. Also, Mtf±Mtg=Mtf±g.
Donsker's Invariance Principle
A discrete random walk can be represented as Sn=ξ1+⋯+ξn. When n becomes very large, the magnitude of the random walk scales by n1. This is due to the Central Limit Theorem - limn→∞Sn/n∼N(0,1). It is a useful fact to note while visualizing the random walk to fit the evolution appropriately in space. Also,
Again, from the CLT, limn→∞Wn(t)→N(0,t). The above formulation takes care of appropriately representing large random walks in a bounded space (probabilistically). Alternately, Wn can be understood as a function from the space C([0,1]→R) - continuous functions from [0,1] to R.
Now, we are interested in performing operations F on these random walks -
n→∞limE[F(Wn)]→E[F(W)]
for all bounded continious functions F:C→R (maps each walk Wn to a real number). Here, W is a standard Brownian motion. All the above equation is saying is that the discretized formulation Wn is equivalent to a standard Brownian motion for practical purposes when n is large.
We have St=St−1⋅ξt where P(ξt=1)=p,P(ξt=−1)=1−p. How do we model this as a difference equation?
St−St−1≡StdSt=St−1(ξt−1)=ξt−1
The RHS should be equated to a⋅dt+b⋅dWt for extending it to random walks. The change in the stock price is represented as a composition of a constant value and a random variable. Interpreting this as an integral, we get the expression
St−S0=∫0tSu(a)du+∫0tSubdWu
In this expression, we know how to evaluate Su(a)du (integrate with respect to time, sum of values as time evolves), but how do we make sense of SubdWu? We evaluate the above expression using Ito's Stochastic Integral.
Intuition - If you think about stock price, it has an inherent drift and randomness. Using this intuition, we get
>dSt=μStdt+σStdWt>
The first integral with respect to du captures the drift whereas the second term captures the randomness.
Ito's Stochastic Integral
How do we calculate the integral ∫0tYsdWs? From Wiener's analysis, we have seen that this expression is a martingale. One approach to evaluate the expression is to it as a combination of Riemman's sums of step functions -
Now, we need to add such Yk's for the required interval (Riemman sum). Ys=∑k=1n1(uk,vk](s)Hk and Hk∈L2(Fuk) -
∫0tYsdWs:=k=1∑n∫0tYs(k)dWs
Since Ys is a continuous martingale in t. Also, the Ito's Isometry property states that
E((∫0tYsdWs)2)=E(∫0tYs2ds)
This is easy to show considering non-overlapping intervals. Hint. Consider conditional expectation using Tower law.
We have seen this in context of Wiener integrals, but the integrand was considered to be deterministic in that case. Here, we generalize the notion by considering integrands that can be stochastic processes as well.
Progressively Measurable - Consider a finite segment of time 0≤t≤T. Then, a stochastic process Xt(ω) can be viewed as a function Ω×[0,T]→R. Then, the integral ∫0tXs(ω)ds can be viewed as a marginal integral of a joint function. The fact that ∫0tXs(ω)ds∈Fs (the integral is measurable) comes from Fubini's theorem in Measure theory. Specifically, the process Xt(ω) is on the domain Ω×[0,t] and is consequently measurable on Ft⊗B[0,t]. Here B refers to Borel Measurability. (Xt) is said to be adapted to (Ft)
Claim. If Xt is adapted and t→Xt(ω) is left-continuous for all ω∈Ω, then X is progressively measurable.
In the previous approach to integrate stochastic processes, we considered Ls class of integrands. That is, we assumed that the integrands can be written as a sequence of step functions. Now, we generalise this notion for any integrands that are progressive measurable
L={Y:Y is progressively measurable and E(∫0TYS2ds)<∞}
A key point to note is that Ls is dense in L (we will not be proving this). That means, given Y∈L, there exists a sequence of integrands (Y(n))n≥1 belonging to Ls such that
n→∞limE[∫0T(Ys−Ys(n))2ds]→0
Since each (Y(n)) is continuous and a martingale, we will try to show that Mt(n)=∫0tYs(n)dWs is a Cauchy sequence converging to the required integral of Ys (note that (Y(n)) is Cauchy) -
since the above is left-continuous, letting n→∞, we get the above expression.
We have derived
Wt2−t=∫0t2WsdWs
We can check Ito's isometry using this.
Also, the above expression in reminiscent of the fundamental theorem of calculus. Using this, we now state the following theorem
Theorem (Ito's Formula/Lemma) For a function f∈C2 and ∣f′(x)∣≤c0ec1∣x∣ for constants c0,c1,
f(Wt)=f(W0)+∫0tf′(Ws)dWs+21∫0tf′′(WS)ds
This forms the basis for the fundamental theorem for stochastic calculus.
Localization
We define the set
Lloc:={Y:Y is progressive and P(∫0tYs2ds<∞)=1}
This is a weaker condition than E[∫0tYS2ds]<∞, and it allows us to generalize the notion of ∫0tYsdWs to more integrands.
Given a Y∈Lloc, we define the stop times τ(n)=inf{t:∫0tYs2ds>n}∧T. Then,
∩n{τ(n)<T}⊂{∫0TYs2ds=∞}
so P(τ(n)<T)→0 as n→∞. Furthermore, defining Ys(n)=1(0,τ(n)](s)Ys, we get
∫0T(Ys(n))2ds≤n2 implying that Y(n)∈L. Also, the martingale ∫0tYs(n)dWs is well-defined for each n=1,2,….
They are also consistent - For m<n, τ(m)<τ(n) and
∫0tYs(m)dWs=∫0tYs(n)dWs
if t∈[0,τ(m)].
Since P(τ(n)<T)→0 as n→∞, and on ∪n{τ(n)=T}, we have
It=nlim∫0tYs(n)dWs,0≤t≤T
and It∧τ(n)=∫0tYs(n)dWs is a martingale.
This process I, which we use as our definition of ∫0tYsdWs is an instance of a local martingale.
A local martingale is an adapted process M=(Mt) for which there is an increasing sequence (σ(n)) of stopping times such that P(σ(n)<T)→0 as n→∞ and Mt∧σ(n) is a martingale.
When the martingales and local martingales have continuous paths, (σ(n)) can be taken to be σ(n)=inf(t:∣Mt∣n)∧T. In particular it can always be arranged that ∣Mt∧σ(n)∣≤Kn for all t.
Ito's Process
An Ito's process is a continuous adapted process X=(Xt) of the form
Xt=X0+∫0tYsdWs+∫0tbsds,0≤t≤T
where Y∈Lloc and b=(bs) is progressive with ∫0T∣bs∣ds<∞.
Revisiting the notion of quadratic variation, the quadratic variation ofX is formally defined as ⟨X⟩ -
⟨X⟩t=∫0tYs2ds,0≤t≤T
For a non-random continuous function g, the process Gt=∫0tg(s)ds has 0 quadratic variation.
Theorem (Ito Formula II) Let f:R×[0,T] be of class C2,1. If X is an Ito's process as defined above, then so is f(Xt,t) with
Note that σWt+μt is a Brownian motion with volatilityσ>0 and driftμ∈R
A trading strategy is a pair ϕ=(α,β) of progressive processes (αt), (βt) such that ∫0Tαs2ds<∞ and ∫0T∣βs∣ds<∞ (they are progressively measurable). α represents the holdings in stock and β represents the holdings in bond. At any time t,
Vt(ϕ)=αtSt+βtBt
For a trading strategy to be self-financing,
Vt(ϕ)−V0(ϕ)=∫0tαudSu+∫0tβudBu,0≤t≤T
where dSu=Su(σdWu+μdu) and dBu=rerudu. In particular, Vt(ϕ) is an Ito process.
The importance of α being progressively measurable comes in handy here, since we want σαSu∈Lloc . That is because Su is positive, ∫0Tαu2Su2du<∞⟺∫0Tαu2du<∞
Suppose we have two Ito's processes Xt=X0+∫0tHsdWs+∫0Tbsds + Yt=Y0+∫0tKsdWs+∫0Tcsds, then their product is also an Ito's process
d(XtYt)=XtdYt+YtdXt+d⟨X,Y⟩t
where ⟨X,Y⟩t=∫0tHsKsds is the quadratic covariation.
Arbitrage
An arbitrage is a self-financing trading strategy ϕ is such that V0(ϕ)=0 and VT(ϕ)≥0 with P(VT(ϕ)>0)>0.
The binomial model is viable and complete. However, it is easy to create arbitrage opportunities in Black-Scholes model -
It⟨I⟩t=∫0tT−s1dWs,0≤t<T=logT−tT↗∞ as t↗T
The quadratic variation represents the energy of the martingale in a way. We further define τq=inf(t:It=a)∧T for some a>0 (P(τa<T)=1). Now, the trading strategy is given by
αtβt=1t≤TT−t1St1=It∧τa−αtSt
Let r=0,Bt=1 for simplicity. Then,
Vt(ϕ)=It∧τa
implying V0(ϕ)=0,VT(ϕ)=Iτa=a>0.
Equivalent Martingale Measure
When μ=0, we have St=S0eσWt−σ2t/2 is a martingale. Then, P itself is an EMM. When μ=0, we need to find a Q on (Ω,FT) that is equivalent to P. How do we check equivalence here? Q(A)>0⟺P(A)>0 for all A∈Ft.
Girsanov's Theorem
Let H∈Lloc and Mt=∫0tHsdWs and ⟨M⟩t=∫0tHs2ds, then
Λt:=exp(Mt−21⟨M⟩t)=1+∫0tΛsdMs
is continuous int, positive, local martingale and a supermartingale. We have E(Λt)≤1, t∈[0,T]. The process is a martingale when E(Λt)=1.
Now, the theorem states that suppose E(Λt)=1 for all t∈[0,T] (⟺E(ΛT)=1⟺Λt is a martingale ). We define
Q(A):=EP(1AΛT),A∈FT
then Q∼P and
Nt:=∫0tYsdWs
for Y∈Lloc is a local martingale with respect to P and
N~t=Nt−∫0tYsHsds
is a local martingale with respect to Q.
Finally, ⟨N⟩t=⟨N~⟩t. In particular, consider
W~t=Wt−∫0tHsds
then W~s is a local martingale with respect to Q and ⟨W~⟩=t. It can be shown that any process with quadratic variation equating to t is a Brownian motion!
Using this theorem, we try and define a martingale for St∗
St∗=S0exp(σWt+(μ−r−σ2/2)t)
Let Ht=θ=σμ−r, then S∗ is a martingale with respect to Q.
By Ito's formula,
St∗dSt∗=σ⎩⎨⎧dWt+θσμ−rdt⎭⎬⎫
Using Girsanov's theorem with Hs=−θ, P∗=ΛtP is an EMM.
where X=(St−K)+,K∗=Ke−rT,Φ= standard normal CDF. Here, f is smooth in the sense that it is twice differentiable in St∗ and differentiable in t. Using Ito's formula, since Vt∗(ϕ) is a martingale,
We can check that when ST>K,αt→1, and ST<K,αt→0.
Let X∗=g(ST∗), then, VT∗(ϕ)=f(St∗,t)=E∗(g(St∗)∣Ft) and limt↑Tf(x,t)=g(x)
Also,
∂t∂f+21σ2x2∂x2∂2f≡0
for (x,t)∈(0,∞)×[0,T). This PDE has a unique solutions when g has certain properties.
Black-Scholes Formula
Consider the market with the assets St=(St0,St1,…,Std) where St0=Bt=e∫0trsds and (rs)0≤s≤T is a progressive process (rs≥0,∫0trsds<∞). This formulation maintains St0dSt0=rtdt.
The stocks have drifts μt1,…,μtd which are progressive processes and ∫0T∣μti∣dt<∞ for all i.
The volatilities are represented as a matrix σT=[σ1σ2…σd] (d×n) where σtij is progressive and ∫0T∥σti∥2dt<∞.
W=(W1,…,Wn) are independent 1-dimensional standard Brownian Motions.
Using these elements, the price process is defined as
Also, given any strategy ϕ~=(ϕ~t1,…,ϕ~td) which is progressive and integrable and c∈R, representing the initial holdings, there exists ϕt0 such that
ϕt=(ϕt0,ϕ~t1,…,ϕ~td) is a self-financing trading strategy V0(ϕ)=c.
A tame trading strategy ϕ there exists c>0 , Vt∗(ϕ)≥−c, ∀t∈[0,T].
Consequently, Vt∗(ϕ)+c⟹Vt∗(ϕ) is a Q-supermartingale.
Proposition - If there is an ELMM Q then no tame trading strategy can be an arbitrage.
Proof. Suppose you have a tame trading strategy with VT∗(ϕ)≥0 and V0∗(ϕ)=0. Due to supermartingale property,
EQ(VT∗(ϕ))≤EQ(V0∗(ϕ))=V0(ϕ)=0
Theorem. (FFTAP)
No free lunch with vanishing risk⟺ there exists an ELMM
ELMM via Girsanov’s theorem - For any Q∼P there exists a Λt, such that dPdQFt=Λt. This is equivalent to saying
Q(A)=EP(1AΛt)
For all A∈Ft. Here, Λt is a P-martingale. We want the form Λt=exp(Mt−21⟨M⟩t)) for the Girsanov’s theorem. Consider the following - Using Ito’s theorem
d(log(Λt))=Λt1dΛt−21Λt21d⟨Λ⟩t
Resembling that dMt=Λt1dΛt.
Using Martingale representation theorem, Mt=−∫0tθs⋅dWs. With Q=exp(MT−21⟨M⟩T)P, Girsanov’s theorem gives us
W~t=Wt+∫0tθsds
is a Q-Brownian motion. Going back to our Black-Scholes model,
We want to choose θ such that W~s captures the drift term. That is, let −σsi⋅θs+μsi−rs=0, then
Sti,∗=S0iexp(∫0tσsi⋅dW~s−21∫0t∣σsi∣2ds)
Simply put, we are trying to find a Q-martingale given σsi,μsi,rs for all ω,s in terms of θs. In other words, we are checking if μs−rs1 is in the column span of σs. We are working our way backwards to find an ELMM Q using the Black-Scholes model. Using θs, we get (Λ)t. But when is this a P-martingale? A super-martingale with same initial and final expectations is a martingale.
Novikov’s condition - If Ep[e21⟨M⟩T]<∞, then Λt=exp(Mt−21⟨M⟩t) for all 0≤t≤T is a P-martingale.
Theorem. If n=d, and σs is invertible for all ω,s and Ep[e21∫0T∣θs∣2ds]<∞, then Q is an ELMM.
With that, we now state the Second Fundamental Theorem of Asset Pricing for continuous models -
Theorem (SFTAP). Suppose there exists an ELMM P∗. Then, the market is complete if the ELMM is unique. By complete, we mean that for all contingent claims X∗∈L1(P∗), there is a self-financing trading strategy such that X=VT(ϕ).
Proof.
(⟹) Let Q be another ELMM, then for A∈FT, we define 1A=X∗∈L1(P∗). That is, X=ST01A (let S00=1).
X∗E∗(VT∗(ϕ))=VT∗(ϕ)=E∗(V0∗(ϕ))=V0(ϕ)=P∗(A)
Also, Vt∗(ϕ),0≤t≤T is a Q-local martingale
Vt∗=V0(ϕ)+∫0tϕsdSs∗
Since the claim is replicable and X∗≥0, Vt∗(ϕ)=E∗(X∗∣Ft)≥0. Since Vt∗(ϕ) is a Q-local martingale, it is also a Q-supermartingale -
P∗(A)=V0(ϕ)≥EQ(VT∗(ϕ))=EQ(X∗)=Q(A)
We have P∗(A)≥Q(A) for all A∈FT. Also, 1−P∗(A)≥1−Q(A) as well, implying P∗=Q.
(⟸) Assume P∗ is the only ELMM. Let X∗∈L1(P∗). The martingale
Mt=E∗(X∗∣FT),0≤t≤T
is right-continuous (continuous). From the martingale representation theorem, there is a unique progressive process ηu=(ηu1,…,ηud) such that
Mt=M0+∫0tηu⋅dSu∗
Now, correspondingly, there exists a unique self-financing trading strategy ϕ=(ϕt0,ηt1,…,ηtd) such that V0(ϕ)=M0 and
where σsi is the ith row of σs which itself is a (d×n) matrix. We typically assume d≤n meaning that we want enough randomness to represent the stocks.
Then, we have the following result -
Lemma. The market is complete iffd=n and σ(w) is invertible
Proof.
(⟸) This is trivial to show
(⟹) Let kt(w)∈n(σt(w)) where n represents the nullspace operation. Also, let ∥kt(w)∥≤1 and ∥kt(w)∥>0 if n(σt(w))={0}. Then, let
X=ST0(1+∫0Tku⋅dW~u)
Since the second term is always a local martingale, and ∥k∥≤1, it's norm is finite and therefore is a martingale. Also, E∗(X∗)=1.
The market being complete means, X∗=1+∫0Tϕu⋅dSu∗ for some trading strategy ϕ. Equating both the equations, we get
implying that the vector kt(w) is in the column space of σt(w)T. For a matrix B, the orthogonal subspace of R(BT) denoted by [R(BT)]⊥=n(B). Therefore, kt(w)⊥n(σt(w)) implying that kt(w)=0!
How do we conclude n=d? If d<n, then the matrix would not be full rank, implying d=n.
Theorem If a market is viable and complete, then the no-arbitrage price of X (assuming X is integrable ∈L1(P∗)) at time t is
Vt(ϕ)=St0E∗(X∗∣Ft)
The proof is similar to the ones we have shown before, and this theorem completes the whole picture of a continuous market.
where α=σT−t and ϕ(z)=2π1e−z2/2. Assume ∣g(y)∣≤C∣y∣p for some p,C>0. That is, we want the contingent claim to have a polynomial growth.
We have (x,t)∈(0,∞)×[0,T), and letting y=xexp(αz−α2/2) we get
∂t∂f+21σ2x2∂x2∂2f=0
Let us delve into an example to understand this better - consider the calculation for E[WT4∣Ft] Note that this process is a martingale. Now, we present two methods to perform this calculation -
Method 1
We say that this expression is equivalent to evaluating a martingale function - =f(Wt,t). We need the property =f(Wt,t)=E[WT4∣Ft], and we choose f(x,t)=x4+6x2(T−t)+3(T−t)2 so that it has the following property -
Therefore, f(Wt,t) is a local martingale. We can check that the integrand fxx=4Ws3+2Ws(T−s) does not integrate to infinite (is a finite value), and consequently, f(Wt,t) is a martingale.
We see that MT=WT4. Let the process Nt=E[WT4∣Ft] - this by itself is also a martingale. Consider the following lemma -
Lemma. Suppose Mt and Nt are martingales with right continuous paths and Mt=NT a.s., then P(Mt=Nt,∀t∈[0,T])=1.
Proof.Mt=E[MT∣Ft]=E[Nt∣Ft]=Nt
This implies that Mt and Nt agree on all rational times (the way we define martingales for Brownian motions). However, since the processes are right continuous , we get
Both boil down to the same sort of an expression that satisfy the heat equation! But how do we guess the f(.,.) in the first method?
Okay, now suppose we have an arbitrary function g and we have to calculate E[g(WT)∣Ft]. We can do the following -
E[g(Wt+(WT−Wt))∣Ft]=E[g(x+zT−t)]∣x=Wt
Since WT−Wt is independent of Ft and Wt is known at Ft, we can remove the conditionality in the expectation!
Then, we say [g(x+T−tx]∣x=Wt=f(Wt,t). For g(x)=x4, we can expand the binomial expression to get f. Furthermore,
f(x,t)=∫−∞∞g(y)pT−t(x,y)dy
where pu(x,y)=2πu1e(y−x)2/2u . This satisfies the heat equation, namely
∂u∂pu=21∂x2∂2pu,∀y
and consequently we can check that the dual is satisfied for f as well!
∂u∂f+21∂x2∂2f=0
resulting in f(Wt,t) always being a martingale!
This works for any g that does not grow too fast (polynomial growth - sufficient not necessary), and it does not have anything to do with smoothness of g either!
Also, limt↑Tf(x,t)=g(x) as long as g is continuous at x.
This indeed is a very powerful result. Let us see how it can be used. Consider another example - Suppose we have μ=r=0, and a σ for a price process St of the form St=S0exp(σWt−σ2t/2). Let the contingent claim be
X=gIT∫0tStdt
That is, the contingent claim is on the total value is a function of the total value accumulated till time T.
So, E[g(It)∣Ft]=ψ(St,It,t) where ψ(x,y,t)=E[g(y+x∫0T−tSs)] if S0=1. The closed form expression depends on g in this case. Again, using Ito's formula on ψ(St,It,t) we get
The expectation of this integral is not trivial. Therefore, we resort to representing the equations as a differential equation. We know Vt=Stϕ(Xt,t) is a martingale from the way it has been constructed. Let us use Ito's formula to represent it. Furthermore, it can be shown that Xt=St−1(K−T1It) is an Ito's process. Why? It is an Ito's process and St−1=S0−1exp(−σWt+σ2t/2) is also an Ito's process - multiplying Ito's processes gives an Ito process. Using Ito's formula -
The stochastic differential equation comes from the fact that Vt is a martingale, and with this we have essentially captured the process in terms of a single spatial random variable. Previously, we had
E(g(IT)∣Ft)=ψ(St,It,t)
where
ψt+xψy+2σ2x2ψxx=0ψ(x,y,T)=(Ty−K)+
In fact, with brutal exercise, we can show that
ψ(x,y,y)=xϕ(xTTK−y,t)
Note that we were able to do such a simplification for a specific nature of g - here g is a straight-line function (gk(cy)=cgk/c(y)). Something similar can be done when g(y)=yp as well.
To show E[XtYt⋅1C]=E[XsYs⋅1C]. Let FtX=σ(XS:s≤t),FtY=σ(Ys:s≤t). Then FtX,Y=σ(FtX,FtY)
This is true for C=A∩B for A∈FsX,B∈FSY. From Monotone Class Theorem, if such a relation is true on the Pi class (closed under intersection), then it is true for all C.
Corollary.⟨X,Y⟩≡0
With these results, we can now work with multiple stocks (d) using the Black-Scholes equation with underlying n-Brownian motion.
Consider this example - -d=1,n=2,r=0. We have dSt=St(βdt+g(Zt)dWt1),S0>0 where dZt=Zt(γdt+δdWt2),Z0=1, β,γ∈R,δ>0,g(x)≥ϵ>0.
Claim. There exists at least one ELMM but this market is not complete.
Proof. Choose a martingale
MTθ:=−∫0tg(Zs)βdWs′+θWt2
for θ∈R. The density process is chosen as Λt=exp(Mtθ−21⟨Mθ⟩t) as suggested by the Girsanov's theorem. Now, we need to show two things - Mtθ is a martingale and Λt satisfied Novikov's condition.
Mtθ is a local martingale because the individual terms in the expression are martingales. The quadratic variation of Mtθ is given by
⟨Mθ⟩t=∫0tg(Zs)2β2ds+θ2t≤ϵ2β2t+θ2t
Since the quadratic variation of Mtθ is bounded by a constant, Mtθ is a martingale. Also, from Novikov's condition, Λt is a martingale as well.
Now, according to Girsanov's theorem,
W~t1=Wt1−⟨W1,Mθ⟩t=Wt1+∫0tg(Zs)βds
is a Qθ-Brownian motion.
From the previous result, W~t2=Wt2−θt is a Q-Brownian motion.
dSt=St(βdt−g(Zt)dW′)=Stg(Zt)dW~t′
So, St is a Qθ-local martingale, implying that Qθ is a ELMM.
Now, to show that the market is not complete, we'll explicitly construct a martingale that cannot be replicated. Letting θ=0, we get W~T2=WT2. To replicate it, we have