Learn more
Learn more
Browse all discussions »
Thank you for enrolling in Introduction to Computational Finance and Financial Econometrics.
Given your interest in this area of study, you may be interested in the other opportunities offered online by the Computational Finance and Risk Management Programs. The Certificate in Computational Finance (http://www.pce.uw.edu/certificates/computational-finance.html) is offered annually and can be completed in three quarters online. Our highly regarded Master’s in Computational Finance and Risk Management is available both online and part-time (http://depts.washington.edu/compfin/content/ms-degree/intro).
Some courses in the certificate programs are available for individual enrollment, and that information can in the course descriptions on our website (http://www.pce.uw.edu/business.html). Feel free to contact a UW enrollment adviser by email at info@pce.uw.edu or by phone at 888-469-6499, if you have additional questions.
Once again, thank you for joining us in this course!
Reminders
Course Calendar ICS
If you use a calendar that accepts .ics files (ex: google, ical), then import the URL below to see due dates for all quizzes and assignments
Upcoming Deadlines
Recent Discussions
Thread title |
|---|
|
Announcements
Week 10 - Introduction to Computational Finance
Congratulations on making it to week 10! This is the final week of the course. We will start with a discussion of portfolio risk budgeting. The goal is to determine the sources of risk in a portfolio due to individual assets. We will learn about marginal contribution to risk and how the beta of an asset can be used as a measure of risk. Next we will introduce Sharpe's Single Index Model. We will study the mathematical properties of this model, as well as estimating the parameters of the model using simple linear regression, and hypothesis testing of these parameters. We will end the course with a discussion of the use of matrix algebra to simplify the use of the Single Index Model with many assets.
There will also be a final exam given this week. The final focuses on the material after the midterm, although concepts from the first half of the course may be incorporated into some of the questions. You will be given 2 attempts for the final with unlimited time. This will hopefully eliminate the technical problems some students experienced with the midterm.
Finally, thank you for taking part in this course! I hope you have found it to be an enjoyable learning experience. It certainly has been for me. If you have found the material interesting, I encourage you to consider some of the other financial courses being offered on Coursera by the University of Washington.
Eric Zivot
There will also be a final exam given this week. The final focuses on the material after the midterm, although concepts from the first half of the course may be incorporated into some of the questions. You will be given 2 attempts for the final with unlimited time. This will hopefully eliminate the technical problems some students experienced with the midterm.
Finally, thank you for taking part in this course! I hope you have found it to be an enjoyable learning experience. It certainly has been for me. If you have found the material interesting, I encourage you to consider some of the other financial courses being offered on Coursera by the University of Washington.
Eric Zivot
Sun 2 Aug 2015 10:00 AM CEST
Introduction to Computational Finance course ending information
On November 4th, Introduction to Computational Finance and Financial Econometrics will be closing for this session.
All quizzes and homework assignments need to be completed and finalized by November 4th. Currently enrolled students can continue to access the course via a “Course Archive” button on the Coursera landing page for the course.
We look forward to continuing to provide you additional courses from the University of Washington and Dr. Eric Zivot in the future.
Thanks you for your participation in this course.
The University of Washington
All quizzes and homework assignments need to be completed and finalized by November 4th. Currently enrolled students can continue to access the course via a “Course Archive” button on the Coursera landing page for the course.
We look forward to continuing to provide you additional courses from the University of Washington and Dr. Eric Zivot in the future.
Thanks you for your participation in this course.
The University of Washington
Sun 2 Aug 2015 10:00 AM CEST
Week 9 - Introduction to Computational Finance
In week 9 we continue our study of portfolio theory. We will consider some examples where we have an arbitrary number of risky assets and use matrix algebra to perform the necessary computations. We will discuss how to use R and Excel for practical computations for finding efficient portfolios. Then we will discuss portfolio theory where we disallow the possibility of short sales. This is an important restriction in practice since many assets are not allowed to be sold short. This restriction complicates the computation of efficient portfolios using matrix algebra and the solution must now be found numerically. We will explore how to do this using the solve.QP() function in R and the Solver in Excel.
Next we will move into a discussion of the statistical analysis of portfolios. It is important to recognize that the inputs to portfolio theory (expected return, standard deviation, covariance, etc) are estimates and are subject to error. This means that the efficient portfolios we compute are also subject to error. We will see how we can use the bootstrap method we discussed earlier to compute standard errors for efficient portfolios. We will conclude this discussion by looking at the stability of efficient portfolios over time.
Eric Zivot
Next we will move into a discussion of the statistical analysis of portfolios. It is important to recognize that the inputs to portfolio theory (expected return, standard deviation, covariance, etc) are estimates and are subject to error. This means that the efficient portfolios we compute are also subject to error. We will see how we can use the bootstrap method we discussed earlier to compute standard errors for efficient portfolios. We will conclude this discussion by looking at the stability of efficient portfolios over time.
Eric Zivot
Sun 26 Jul 2015 10:00 AM CEST
Week 8 - Introduction to Computational Finance
In week 8 we begin our study of portfolio theory. Portfolio theory is the study of optimal asset allocation. We will start with a simple two risky asset example and use this to illustrate the risk and return characteristics of portfolios. We will learn about efficient portfolios: those portfolios with the highest expected return for a given level of risk (volatility). We will learn how to construct a frontier of all the efficient portfolios, and from that determine the best portfolio for a given investor based on their risk preferences.
Next, we will discuss portfolios with one risky asset and one risk free asset. Risk free assets are used quite heavily in the real world and its important to understand how a risk free asset influences portfolios. We will study the risk-return tradeoffs in this context and learn about the Sharpe Ratio. We will then extend the discussion to include two risky assets and a risk free asset. Here we will see that the set of efficient portfolios are combinations of the risk free asset and something called the tangency portfolio. The tangency portfolio is the portfolio of two risky assets with the highest Sharpe Ratio.
Finally, we will see how matrix algebra allows us to extend the computations for two risky assets to the general case of an arbitrary number of risky assets.
Eric Zivot
Next, we will discuss portfolios with one risky asset and one risk free asset. Risk free assets are used quite heavily in the real world and its important to understand how a risk free asset influences portfolios. We will study the risk-return tradeoffs in this context and learn about the Sharpe Ratio. We will then extend the discussion to include two risky assets and a risk free asset. Here we will see that the set of efficient portfolios are combinations of the risk free asset and something called the tangency portfolio. The tangency portfolio is the portfolio of two risky assets with the highest Sharpe Ratio.
Finally, we will see how matrix algebra allows us to extend the computations for two risky assets to the general case of an arbitrary number of risky assets.
Eric Zivot
Sun 19 Jul 2015 10:00 AM CEST
Week 7 - Introduction to Computational Finance
Week 7 - Introduction to Computational Finance,
In week 7, we finish our discussion of estimation theory with a practical problem of estimating Value at Risk (VaR) in the constant expected return model. We will see that it is quite difficult to derive an analytic formula for the standard error of VaR which will motivate our discussion of the Bootstrap method. This is a computer simulation technique that will allow us to compute standard errors without an analytical expression. You will see that is is closely related to Monte Carlo simulation that we learned about last week.
Next, we will finish our review of statistical estimation theory with a discussion of hypothesis testing. Once we have an estimated model, we would like to test hypotheses about the model. For example, we may want to test the hypothesis that the mean of the model is 0. Alternatively, we can use hypothesis testing to verify or refute assumptions in the model, such as are the returns normally distributed? We will review the concepts of hypothesis testing and learn about test statistics. Finally, we will discuss applications of hypothesis testing to the constant expected return model.
Eric Zivot
In week 7, we finish our discussion of estimation theory with a practical problem of estimating Value at Risk (VaR) in the constant expected return model. We will see that it is quite difficult to derive an analytic formula for the standard error of VaR which will motivate our discussion of the Bootstrap method. This is a computer simulation technique that will allow us to compute standard errors without an analytical expression. You will see that is is closely related to Monte Carlo simulation that we learned about last week.
Next, we will finish our review of statistical estimation theory with a discussion of hypothesis testing. Once we have an estimated model, we would like to test hypotheses about the model. For example, we may want to test the hypothesis that the mean of the model is 0. Alternatively, we can use hypothesis testing to verify or refute assumptions in the model, such as are the returns normally distributed? We will review the concepts of hypothesis testing and learn about test statistics. Finally, we will discuss applications of hypothesis testing to the constant expected return model.
Eric Zivot
Sun 12 Jul 2015 10:00 AM CEST
Week 6 - Introduction to Computational Finance
Welcome to week 6 - Introduction to Computational Finance,
In week 6 we discuss our first formal model for asset returns, the constant expected return model. We begin by reviewing the assumptions of the model and then deriving some mathematical properties for the model. We will learn how to simulate samples from our proposed model by introducing a very powerful and commonly used technique called Monte Carlo simulation. If the simulated data looks like actual data, it provides us with a first step validation of the model. In addition, as models typically rely on unknown parameters that need to be estimated or calibrated to actual data, our discussion includes a of review statistical estimation theory. We will learn about the properties of estimators and how to quantify the estimation error using standard errors and confidence intervals. Finally, we will see how we can propose estimators for parameters based on some the sample statistics we used in previous weeks.
There is a midterm exam for week 6. You can access the midterm in the "Exams" section on the left of the course page. The midterm includes material through week 5 on descriptive statistics. The questions are very similar to those found in the homework assignments. You will have 1 hour and 30 minutes to complete the exam, and are allowed only one submission. The timer starts as soon as you access the midterm, so please do not attempt it until you have the free time to do so. You should use the midterm as a form of self-assessment in order to gauge your understanding of the material through week 5.
Eric Zivot
In week 6 we discuss our first formal model for asset returns, the constant expected return model. We begin by reviewing the assumptions of the model and then deriving some mathematical properties for the model. We will learn how to simulate samples from our proposed model by introducing a very powerful and commonly used technique called Monte Carlo simulation. If the simulated data looks like actual data, it provides us with a first step validation of the model. In addition, as models typically rely on unknown parameters that need to be estimated or calibrated to actual data, our discussion includes a of review statistical estimation theory. We will learn about the properties of estimators and how to quantify the estimation error using standard errors and confidence intervals. Finally, we will see how we can propose estimators for parameters based on some the sample statistics we used in previous weeks.
There is a midterm exam for week 6. You can access the midterm in the "Exams" section on the left of the course page. The midterm includes material through week 5 on descriptive statistics. The questions are very similar to those found in the homework assignments. You will have 1 hour and 30 minutes to complete the exam, and are allowed only one submission. The timer starts as soon as you access the midterm, so please do not attempt it until you have the free time to do so. You should use the midterm as a form of self-assessment in order to gauge your understanding of the material through week 5.
Eric Zivot
Sun 5 Jul 2015 10:00 AM CEST
Week 5 - Introduction to Computational Finance
Welcome to week 5 - Introduction to Computational Finance,
Congratulations on making it to the midpoint of the course! In week 5 we begin our discussion of descriptive statistics for asset returns. We will look at actual monthly and daily continuously compounded returns for a number of representative stocks and portfolios . Our goal is to develop some stylized facts about the distribution of asset returns that will help us develop formal models later on in the course. We will start with univariate descriptive statistics, considering graphical diagnostics such as the histogram, box-plots, QQ plots, and the empirical cumulative distribution function. We will also look at numerical descriptive statistics such as the sample mean, standard deviation, skewness and kurtosis. We will then move on to bivariate descriptive statistics to help summarize the relationship and dependence between two asset returns. We will look at graphical measures such as scatterplots and numerical properties such as the sample covariance and correlation. Finally, we will consider time series descriptive statistics such as the sample autocorrelation.
There will be a midterm exam given next week. It will cover all the material through week 5, with questions very similar to those given in the homework assignments. However, unlike the homework assignments the midterm will be timed and you will only be allowed one try. You should use the midterm as a way to self-assess your progress and understanding of the course material. I will release more details about the midterm in an email and announcement next week.
Eric Zivot
Congratulations on making it to the midpoint of the course! In week 5 we begin our discussion of descriptive statistics for asset returns. We will look at actual monthly and daily continuously compounded returns for a number of representative stocks and portfolios . Our goal is to develop some stylized facts about the distribution of asset returns that will help us develop formal models later on in the course. We will start with univariate descriptive statistics, considering graphical diagnostics such as the histogram, box-plots, QQ plots, and the empirical cumulative distribution function. We will also look at numerical descriptive statistics such as the sample mean, standard deviation, skewness and kurtosis. We will then move on to bivariate descriptive statistics to help summarize the relationship and dependence between two asset returns. We will look at graphical measures such as scatterplots and numerical properties such as the sample covariance and correlation. Finally, we will consider time series descriptive statistics such as the sample autocorrelation.
There will be a midterm exam given next week. It will cover all the material through week 5, with questions very similar to those given in the homework assignments. However, unlike the homework assignments the midterm will be timed and you will only be allowed one try. You should use the midterm as a way to self-assess your progress and understanding of the course material. I will release more details about the midterm in an email and announcement next week.
Eric Zivot
Sun 28 Jun 2015 10:00 AM CEST
Week 4 - Introduction to Computational Finance
Today we begin week 4 of the course.
We start with an example of portfolio math using matrix algebra. You will see how matrix algebra simplifies many of these computations. We then conclude our review with a discussion on the multivariate normal distribution. Again, you will see how matrix algebra can simplify how we characterize these distributions.
Now that we've gotten the probability and matrix algebra reviews under our belts, we can shift our focus towards the analysis of financial data. Our goal is to analyze asset returns over time; in order to properly characterize such data; we need to learn about the properties of time series. We will devote the rest of week 4 to a discussion of time series concepts. Some of the topics we will cover include stochastic processes, stationarity, autocorrelation, and the moving average and autoregressive processes.
Eric Zivot
We start with an example of portfolio math using matrix algebra. You will see how matrix algebra simplifies many of these computations. We then conclude our review with a discussion on the multivariate normal distribution. Again, you will see how matrix algebra can simplify how we characterize these distributions.
Now that we've gotten the probability and matrix algebra reviews under our belts, we can shift our focus towards the analysis of financial data. Our goal is to analyze asset returns over time; in order to properly characterize such data; we need to learn about the properties of time series. We will devote the rest of week 4 to a discussion of time series concepts. Some of the topics we will cover include stochastic processes, stationarity, autocorrelation, and the moving average and autoregressive processes.
Eric Zivot
Sun 21 Jun 2015 10:00 AM CEST
Week 3 - Introduction to Computational Finance
Welcome to week 3 of the course.
We will start by concluding our review of probability theory. We will learn about bivariate distributions and the important concepts of covariance and correlation. These two concepts are used very heavily in finance. We will also discuss the linear combination of random variables and how it relates to the analysis of portfolios. We end the week with an introduction to matrix algebra. Matrix algebra is especially important since many of the computations we will do involving more than two assets are greatly simplified by using matrix algebra. If you are a bit rusty on your matrix algebra, or have never taken a linear algebra course before, this is a good time to brush up! Furthermore, don't forget to complete the R programming assignments on DataCamp.
Eric Zivot
We will start by concluding our review of probability theory. We will learn about bivariate distributions and the important concepts of covariance and correlation. These two concepts are used very heavily in finance. We will also discuss the linear combination of random variables and how it relates to the analysis of portfolios. We end the week with an introduction to matrix algebra. Matrix algebra is especially important since many of the computations we will do involving more than two assets are greatly simplified by using matrix algebra. If you are a bit rusty on your matrix algebra, or have never taken a linear algebra course before, this is a good time to brush up! Furthermore, don't forget to complete the R programming assignments on DataCamp.
Eric Zivot
Sun 14 Jun 2015 10:00 AM CEST
Week 2 - Introduction to Computational Finance
Welcome to week 2 - Introduction to Computational Finance,
I hope that you enjoyed the first week on return calculations and are becoming familiar with using R. We are aware of the discussion about the quality of the video in the lectures. Please note that the intended and optimum way for you to view the lectures is to stream them through the Coursera website. This allows you to view both the video and lecture slides at the same time, and to dynamically switch between them. Instructions on how to use the technology are available on the Viewing the Video Lectures course page.
This week we begin our review of probability theory. We will learn about random variables and distribution functions for discrete and continuous random variables. Particular attention will be paid to the normal distribution and its use in financial modeling. We will also discuss the shape characteristics of distributions such as expected value, standard deviation, skewness and kurtosis. Finally, we define the risk concept, Value-at-Risk, and how it relates to the quantiles of a distribution. These probability concepts will serve as a foundation for the rest of the course.
Eric Zivot
I hope that you enjoyed the first week on return calculations and are becoming familiar with using R. We are aware of the discussion about the quality of the video in the lectures. Please note that the intended and optimum way for you to view the lectures is to stream them through the Coursera website. This allows you to view both the video and lecture slides at the same time, and to dynamically switch between them. Instructions on how to use the technology are available on the Viewing the Video Lectures course page.
This week we begin our review of probability theory. We will learn about random variables and distribution functions for discrete and continuous random variables. Particular attention will be paid to the normal distribution and its use in financial modeling. We will also discuss the shape characteristics of distributions such as expected value, standard deviation, skewness and kurtosis. Finally, we define the risk concept, Value-at-Risk, and how it relates to the quantiles of a distribution. These probability concepts will serve as a foundation for the rest of the course.
Eric Zivot
Sun 7 Jun 2015 10:00 AM CEST
Welcome to Computational Finance and Financial Econometrics!
Hello, and welcome to Computational Finance and Financial Econometrics!
I am Eric Zivot, Professor in the Economics Department at the University of Washington, Adjunct Professor of Statistics, Adjunct Professor of Finance, and Adjunct Professor of Applied Mathematics. I am very pleased to welcome you to Computational Finance and Financial Econometrics.
My goals for you in this course are that you will be able to move from an economic viewpoint to an econometric model that works with real-world data. I look forward to seeing what new knowledge will rise out of your collaboration with this huge and diverse community of fellow learners.
I suggest you begin by going to discussion forums and introducing yourself. You should then orient yourself to the course format by visiting the "Welcome to Computational Finance and Financial Econometrics!" page, and viewing the introductory video (if you haven't already done so).
Again, welcome to the course!
Eric
I am Eric Zivot, Professor in the Economics Department at the University of Washington, Adjunct Professor of Statistics, Adjunct Professor of Finance, and Adjunct Professor of Applied Mathematics. I am very pleased to welcome you to Computational Finance and Financial Econometrics.
My goals for you in this course are that you will be able to move from an economic viewpoint to an econometric model that works with real-world data. I look forward to seeing what new knowledge will rise out of your collaboration with this huge and diverse community of fellow learners.
I suggest you begin by going to discussion forums and introducing yourself. You should then orient yourself to the course format by visiting the "Welcome to Computational Finance and Financial Econometrics!" page, and viewing the introductory video (if you haven't already done so).
Again, welcome to the course!
Eric
Mon 1 Jun 2015 10:00 AM CEST
University of Washington Certificate Programs
Thank you for enrolling in Introduction to Computational Finance and Financial Econometrics.
Given your interest in this area of study, you may be interested in the other opportunities offered online by the Computational Finance and Risk Management Programs. The Certificate in Computational Finance (http://www.pce.uw.edu/certificates/computational-finance.html) is offered annually and can be completed in three quarters online. Our highly regarded Master’s in Computational Finance and Risk Management is available both online and part-time (http://depts.washington.edu/compfin/content/ms-degree/intro).
Some courses in the certificate programs are available for individual enrollment, and that information can in the course descriptions on our website (http://www.pce.uw.edu/business.html). Feel free to contact a UW enrollment adviser by email at info@pce.uw.edu or by phone at 888-469-6499, if you have additional questions.
Once again, thank you for joining us in this course!
Mon 1 Jun 2015 10:00 AM CEST
R Programming Assignments with DataCamp
Hi,
Throughout the course lectures I explain how to use the open source statistical programming language R to analyze financial data, estimate statistical models, and construct optimized portfolios. However, since theory is only half the story, and practice makes perfect, I complemented these lectures with weekly programming assignments that will help you to better understand the covered material. These programming assignments are to be completed on the web-based platform DataCamp, and every week a link to the new assignment will be posted on the course page. You can attempt these assignments as many times as you like.
DataCamp is an online interactive learning platform that offers free R tutorials through learning-by-doing. Since R has a rather steep learning curve, DataCamp is a perfect initiative to help you understand the underlying logic of the R code explained in the lectures. It breaks the different programming assignments down into many short exercises, and provides you with hints and instant feedback on how to perform even better. This helps you to understand deeply what is going on, even if you are a novice to programming.
If you have any questions or feedback on the platform itself, you can always send them a message via Facebook or Twitter.
Hope you enjoy the interactive R assignments!
Eric
Throughout the course lectures I explain how to use the open source statistical programming language R to analyze financial data, estimate statistical models, and construct optimized portfolios. However, since theory is only half the story, and practice makes perfect, I complemented these lectures with weekly programming assignments that will help you to better understand the covered material. These programming assignments are to be completed on the web-based platform DataCamp, and every week a link to the new assignment will be posted on the course page. You can attempt these assignments as many times as you like.
DataCamp is an online interactive learning platform that offers free R tutorials through learning-by-doing. Since R has a rather steep learning curve, DataCamp is a perfect initiative to help you understand the underlying logic of the R code explained in the lectures. It breaks the different programming assignments down into many short exercises, and provides you with hints and instant feedback on how to perform even better. This helps you to understand deeply what is going on, even if you are a novice to programming.
If you have any questions or feedback on the platform itself, you can always send them a message via Facebook or Twitter.
Hope you enjoy the interactive R assignments!
Eric
Mon 1 Jun 2015 10:00 AM CEST