betaFeedback

WorksEconomics, Econometrics and Finance

An Introduction to Financial Derivatives: Open Lecture Notes

Abstract

These lecture notes give a broad introduction to financial derivatives at the level of a Masters program in quantitative finance. Topics include forward and futures contracts, no-arbitrage bounds and put-call parity, the binomial model, the fundamental theorems of asset pricing, Brownian motion and Itô calculus, the Black-Scholes model, the Greeks, variance swaps, local and stochastic volatility models including Heston, and static hedging of barrier options. Each chapter ends with exercises, and complete solutions, including Python code for the computational exercises, are provided in an appendix. The notes are open source: the LaTeX source is available at https://github.com/mattlorig/intro-financial-derivatives, and readers are invited to submit corrections and additions.

© 2026 Matthew Lorig.

This work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0), https://creativecommons.org/licenses/by-sa/4.0/. You are free to share (copy and redistribute) and adapt (remix, transform and build upon) this material for any purpose, including commercially, provided that you (i) give appropriate credit to the original author, (ii) indicate if changes were made, and (iii) distribute your contributions under the same license. Code listings in these notes are additionally released under the MIT License.

The LaTeX source of these notes is available at https://github.com/mattlorig/intro-financial-derivatives.

Please cite as: M. Lorig, An Introduction to Financial Derivatives: Open Lecture Notes, https://github.com/mattlorig/intro-financial-derivatives, 2026.

Preface

These notes give a broad introduction to financial derivatives. They were developed for CFRM 504, a course for Masters students in Computational Finance and Risk Management at the University of Washington, and they assume a working knowledge of calculus and basic probability. Because the focus of this course is on the applied aspects of this topic, we will sometimes forgo mathematical rigor, favoring instead a heuristic development. The mathematical statements in these notes should be taken as “true in spirit,” but perhaps not always rigorously true mathematically. The hope is that, what the notes lack in rigor, they make up in clarity.

These notes are very much a work in progress.

How to contribute

These notes are open source. The LaTeX source is available at https://github.com/mattlorig/intro-financial-derivatives, and the entire document compiles from that single file, with no external figures or bibliography files. Anyone may download it, correct errors, add examples, exercises or sections, and adapt the notes for their own courses, subject to the license on the reverse of the title page. To report an error, open an issue on the repository; to propose a change, submit a pull request. Contributors whose changes are accepted will be listed below.

Contributors

The author of these notes wishes to express his sincere thanks to the following people for their contributions.

  • Weston Barger: checking and writing homework solutions, as well as making corrections and improvements to the text.

Donations

If you find these notes useful and would like to make a donation to help me develop them further you can donate ETH or ERC20 tokens to the address below.

ETH address: 0x3D6042D6a6f39735bbAd43ad1Aa676113D55BaAC

Review of Probability

In this chapter, we will very briefly review some concepts from probability theory. This material is essential for understanding the subsequent chapters.

Sample spaces, events, and probability measures

We begin this section with some important definitions.

Definition 1.1.1. The set of all possible outcomes of a random experiment is called the sample space and is denoted by Ω\Omega.

An outcome of a random experiment is an element of Ω\Omega and is denoted by ω\omega.

Definition 1.1.2. An event is any subset of the sample space Ω\Omega. We denote events by capital roman letters A, B, C, . . ..

Finally, we will denote by F\mathcal{F} all of the subsets of a sample space Ω\Omega.1 Note that a single element ω\omega of Ω\Omega is also a subset of Ω\Omega, as are the empty set ∅\emptyset and Ω\Omega itself. Thus, we have {ω},∅,Ω∈F\{\omega\}, \emptyset, \Omega\in\mathcal{F}.

Example 1.1.3 (Toss a coin). When one tosses a coin, there are two possible outcomes: heads (denoted by H) and tails (denoted by T). Thus, our sample space is Ω={H,T}\Omega= \{H, T\}. One possible element or outcome of Ω\Omega is to “toss a heads” ω=H\omega= H. Two possible events are “toss a heads” A={H}A = \{H\} and “toss a heads or a tails” B={H,T}B = \{H, T\}.

Example 1.1.4 (Roll a die). A standard die has six sides, labeled one through six. Thus, our sample space is Ω={1,2,3,4,5,6}\Omega= \{1, 2, 3, 4, 5, 6\}. One possible element or outcome of Ω\Omega is “roll a two” ω=2\omega= 2. Two possible events are “roll an odd number” A={1,3,5}A = \{1, 3, 5\} and “roll a one or two” B={1,2}B = \{1, 2\}.

So far, we have not yet talked about probabilities – only outcomes of a random experiment (elements ω∈Ω\omega\in\Omega) and events (subsets A∈FA \in\mathcal{F}). We now introduce the concept of a probability measure.

Definition 1.1.5. A probability measure P\mathbb{P} assigns probabilities to events P:F→[0,1]\mathbb{P}:\mathcal{F}\to[0,1] and satisfies

  1. P(Ω)=1\mathbb{P}(\Omega)=1, and

  1. If (Ai)(A_i) is a countable sequence of disjoint subsets of Ω\Omega, (i.e., Ai∩Aj=∅A_i\cap A_j=\varnothing if i≠ji\ne j), then

P(⋃iAi)=∑iP(Ai).\mathbb{P}\left(\bigcup_i A_i\right)=\sum_i\mathbb{P}(A_i).

The first property in Definition 1.1.5 is easy to understand. When we perform a random experiment, the probability that something happens is one. Thus, in order for a probability measure P\mathbb{P} to make any sense at all, we must have P(Ω)=1\mathbb{P}(\Omega)=1. The second property in Definition 1.1.5, known as countable additivity, ensures that we do not double-count the probability of events. For example, if we toss a coin, the probability of getting a heads or tails should equal the probability of getting a heads plus the probability of getting a tails

P(H∪T)=P(H)+P(T).\mathbb{P}(\mathrm{H}\cup\mathrm{T})=\mathbb{P}(\mathrm{H})+\mathbb{P}(\mathrm{T}).

Note that, together, the two properties in Definition 1.1.5 imply that P(∅)=0\mathbb{P}(\varnothing)=0 as

1=P(Ω)=P(Ω∪∅)=P(Ω)+P(∅)=1+P(∅),⇒P(∅)=0,1=\mathbb{P}(\Omega)=\mathbb{P}(\Omega\cup\varnothing)=\mathbb{P}(\Omega)+\mathbb{P}(\varnothing)=1+\mathbb{P}(\varnothing),\qquad\Rightarrow\qquad\mathbb{P}(\varnothing)=0,

where we have used the fact that Ω\Omega and ∅\varnothing are disjoint. As we shall see with the next few examples, a probability measure P\mathbb{P} need not assign probabilities to events in a way that is consistent with our real-world experience.

Example 1.1.6 (Toss a coin). When we toss a coin, our experience tells us that the probability of getting a heads is 1/21/2 and the probability of getting tails is 1/21/2. Thus, if we want a probability measure P\mathbb{P} to reflect our experience, we should have P(H)=P(T)=1/2\mathbb{P}(\mathrm{H})=\mathbb{P}(\mathrm{T})=1/2. We can think of P\mathbb{P} as the “real world” or “statistical” probability measure. However, we could consider another probability measure P~\widetilde{\mathbb{P}} that assigns different probabilities to heads and tails. For example, we could have P~(H)=3/4\widetilde{\mathbb{P}}(\mathrm{H})=3/4 and P~(T)=1/4\widetilde{\mathbb{P}}(\mathrm{T})=1/4. The measure P~\widetilde{\mathbb{P}} does not capture what goes on in the real world. But, it is nonetheless a valid probability measure for the random experiment of tossing a coin. You should check for yourself that P~\widetilde{\mathbb{P}} satisfies the conditions of Definition 1.1.5.

Example 1.1.7 (Roll a die). When we roll a die, our experience tells us that the probability of getting any number ω∈{1,2,3,4,5,6}\omega\in\{1,2,3,4,5,6\} is 1/61/6. Thus, if we want a probability measure P\mathbb{P} to reflect our experience,

we should have P(ω)=1/6\mathbb{P}(\omega) = 1/6 for all ω∈{1,2,3,4,5,6}\omega\in\{1, 2, 3, 4, 5, 6\}. As in the coin toss example, we can think of P\mathbb{P} as the “real world” or “statistical” probability measure. However, we could consider another probability measure P~\widetilde{\mathbb{P}} that assigns different probabilities to each outcome. For example, we could have P~(1)=1/2\widetilde{\mathbb{P}}(1) = 1/2 and P~(ω)=1/10\widetilde{\mathbb{P}}(\omega) = 1/10 for ω∈{2,3,4,5,6}\omega\in\{2, 3, 4, 5, 6\}. The measure P~\widetilde{\mathbb{P}} does not capture what goes on in the real world. But, it is nonetheless a valid probability measure for the random experiment of rolling a die. You should check for yourself that P~\widetilde{\mathbb{P}} satisfies the conditions of Definition 1.1.5.

It may not be clear at this time why one would ever consider a probability measure P~\widetilde{\mathbb{P}} that does not reflect our real-world experience. However, when we discuss option pricing, it will become apparent how such a measure can be useful.

Definition 1.1.8. Two probability measures P\mathbb{P} and P~\widetilde{\mathbb{P}} are said to be equivalent, denoted P∼P~\mathbb{P} \sim\widetilde{\mathbb{P}}, if they agree on events that occur with probability one. That is, if

P(A)=1⟺P~(A)=1.\mathbb{P}(A) = 1 \quad\Longleftrightarrow\quad\widetilde{\mathbb{P}}(A) = 1.

Note that if P∼P~\mathbb{P} \sim\widetilde{\mathbb{P}} we also have

P(A)=0⟺P~(A)=0.\mathbb{P}(A) = 0 \quad\Longleftrightarrow\quad\widetilde{\mathbb{P}}(A) = 0.

In other words, equivalent measures agree on events that occur with probability one and on events that occur with probability zero.

Example 1.1.9. Let us return to the coin toss example. Consider two probability measures P\mathbb{P} and P~\widetilde{\mathbb{P}}. Suppose that P(H)=1/2\mathbb{P}(\mathrm{H}) = 1/2 and P~(H)=3/4\widetilde{\mathbb{P}}(\mathrm{H}) = 3/4. Then P∼P~\mathbb{P} \sim\widetilde{\mathbb{P}} because

P(H∪T)=P(H)+P(T)=1/2+1/2=1,\mathbb{P}(\mathrm{H} \cup\mathrm{T}) = \mathbb{P}(\mathrm{H}) + \mathbb{P}(\mathrm{T}) = 1/2 + 1/2 = 1,
P~(H∪T)=P~(H)+P~(T)=3/4+1/4=1.\widetilde{\mathbb{P}}(\mathrm{H} \cup\mathrm{T}) = \widetilde{\mathbb{P}}(\mathrm{H}) + \widetilde{\mathbb{P}}(\mathrm{T}) = 3/4 + 1/4 = 1.

Now, consider a third probability measure P^\widehat{\mathbb{P}} that satisfies P^(H)=1\widehat{\mathbb{P}}(\mathrm{H}) = 1. Then P^\widehat{\mathbb{P}} is not equivalent to P\mathbb{P} (or to P~\widetilde{\mathbb{P}}) because P(H)=1/2≠1\mathbb{P}(\mathrm{H}) = 1/2 \ne1.

Random variables

Intuitively, we think of a random variables as a quantity whose value is uncertain. For example, the value of a stock 10 days from now is uncertain and, as such, we think of it as being random. We give a precise definition of a random variable below.

Definition 1.2.1. A random variable XX is a function that maps every outcome ω\omega of a random experiment to a number in R\mathbb{R}

X:Ω→R.X:\Omega\to\mathbb{R}.

The set SS of values that XX can take is called the state space of XX.

Example 1.2.2 (Toss a coin). If we toss a coin, we have Ω={H,T}\Omega=\{\mathrm{H},\mathrm{T}\}. Two possible random variables on this sample space are

X(ω)={1,if ω=H,0,if ω=T,Y(ω)={10,if ω=H,20,if ω=T.X(\omega)= \begin{cases} 1, & \text{if }\omega=\mathrm{H},\\ 0, & \text{if }\omega=\mathrm{T}, \end{cases} \qquad Y(\omega)= \begin{cases} 10, & \text{if }\omega=\mathrm{H},\\ 20, & \text{if }\omega=\mathrm{T}. \end{cases}

Thus, if we toss a heads ω=H\omega=\mathrm{H}, we have X=1X=1 and Y=10Y=10. The state space of XX is S={0,1}S=\{0,1\}. The state space of YY is S={10,20}S=\{10,20\}.

Example 1.2.3 (Roll a die). If we roll a die, we have Ω={1,2,3,4,5,6}\Omega=\{1,2,3,4,5,6\}. Two possible random variables on this sample space are

X(ω)=ω,Y(ω)=ω2.X(\omega)=\omega,\qquad Y(\omega)=\omega^2.

Thus, if we roll a three ω=3\omega=3, we have X=3X=3 and Y=32=9Y=3^2=9. The state space of XX is S={1,2,3,4,5,6}S=\{1,2,3,4,5,6\}. The state space of YY is S={1,4,9,16,25,36}S=\{1,4,9,16,25,36\}.

Random variables can be described by their cumulative distribution function.

Definition 1.2.4. The cumulative distribution function (CDF) of a random variable XX is a function FX:R→[0,1]F_X:\mathbb{R}\to[0,1] defined by

FX(x):=P(X≤x),(1)F_X(x):=\mathbb{P}(X\le x), \tag*{(1)}

where x∈Rx\in\mathbb{R}.

Here, we have used the short-hand notation

{X≤x}={ω∈Ω:X(ω)≤x}.\{X\le x\}=\{\omega\in\Omega:X(\omega)\le x\}.

Now, we make an important note. A random variable XX is defined independently of a probability measure. However, we see from (1) that the CDF FXF_X is defined with respect to a probability measure P\mathbb{P}. Thus a random variable XX can have a CDF FXF_X under P\mathbb{P} and have an entirely different CDF F~X\widetilde{F}_X under P~\widetilde{\mathbb{P}}. We will (hopefully) make this concept clear through some examples in the coming sections.

Sometimes, the value of one random variable XX may affect the value of another random variable YY. For example, suppose you roll two 6-sided dice. Let XX be the result of the first roll, let YY be the result of the second and let Z=X+YZ=X+Y. If I tell you that Z=2Z=2, then you know that both X=1X=1 and Y=1Y=1. In cases such as these, in order to describe the relationship between two random variables, we will need to consider joint cumulative distribution function.

Definition 1.2.5. The joint cumulative distribution function (joint CDF) of two random variables XX and YY is a function FX,Y:R2→[0,1]F_{X,Y}:\mathbb{R}^{2}\to[0,1] defined by

FX,Y(x,y):=P(X≤x,Y≤y),(2)F_{X,Y}(x,y):=\mathbb{P}(X\le x,Y\le y), \tag*{(2)}

where x,y∈Rx,y\in\mathbb{R}.

Here, we have used the short-hand notation

{X≤x,Y≤y}={ω∈Ω:X(ω)≤x,Y(ω)≤y}.\{X\le x,Y\le y\}=\{\omega\in\Omega:X(\omega)\le x,Y(\omega)\le y\}.

As in the univariate case, the joint CDF FX,YF_{X,Y} is defined with respect to a particular probability measure (in the case of equation (2), the probability measure is P\mathbb{P}).

Definition 1.2.5 can easily be extended to any finite number of random variables. For example, the joint CDF of three random variables (X,Y,Z)(X,Y,Z) would be defined as follows

References

  1. [1]Carr, P., & Madan, D. (1998). Towards a theory of volatility trading. In Volatility: New Estimation Techniques for Pricing Derivatives (pp. 417–427).
  2. [2]Heston, S. L. (1993). A closed-form solution for options with stochastic volatility with applications to bond and currency options. The Review of Financial Studies 6(2), 327–343.DOI

Paper details

vlnc:W_caf27p7n
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY-SA 4.0
Usage

views · 7

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Economics, Econometrics and Finance

Not recorded for this paper

View statement
Looking for related papers…
Contents