Book Image

Python for Finance - Second Edition

By : Yuxing Yan
5 (1)
Book Image

Python for Finance - Second Edition

5 (1)
By: Yuxing Yan

Overview of this book

This book uses Python as its computational tool. Since Python is free, any school or organization can download and use it. This book is organized according to various finance subjects. In other words, the first edition focuses more on Python, while the second edition is truly trying to apply Python to finance. The book starts by explaining topics exclusively related to Python. Then we deal with critical parts of Python, explaining concepts such as time value of money stock and bond evaluations, capital asset pricing model, multi-factor models, time series analysis, portfolio theory, options and futures. This book will help us to learn or review the basics of quantitative finance and apply Python to solve various problems, such as estimating IBM’s market risk, running a Fama-French 3-factor, 5-factor, or Fama-French-Carhart 4 factor model, estimating the VaR of a 5-stock portfolio, estimating the optimal portfolio, and constructing the efficient frontier for a 20-stock portfolio with real-world stock, and with Monte Carlo Simulation. Later, we will also learn how to replicate the famous Black-Scholes-Merton option model and how to price exotic options such as the average price call option.
Table of Contents (23 chapters)
Python for Finance Second Edition
Credits
About the Author
About the Reviewers
www.PacktPub.com
Customer Feedback
Preface
Index

Skewness and kurtosis


Based on the normality assumption, a VaR estimation considers only the first two moments: mean and variance. If stock returns truly follow a normal distribution, those two moments would fully define their probability distribution. From the preceding sections, we know that this is not true. The first remedy is to include other higher moments in addition to the first two moments. The third and fourth moments are called skewness and kurtosis. For a stock or portfolio with n returns, skewness is estimated by the following formula:

Here, skewness is the skewness, Ri is the ith return, is the mean return, n is the number of returns, and σ is the standard deviation of returns. The kurtosis reflects the impact of extreme values because a power of 4 is very high. The kurtosis is usually estimated by the following formula is:

For a standard moral distribution, it has a zero mean, unit variance, zero skewness, and its kurtosis is 3. Because of this, sometimes kurtosis is defined...