Tuesday, November 20, 2012

Online Courses, Virtual Trading Rooms

I just returned from a conference where there was a panel discussion on university trading rooms. One point of discussion concerned online and distance education: how can students who are not on campus benefit from a trading room? The question becomes even more important given the large increase in online courses; according to the 2010 Sloan Survey of Online Education, online enrollment increased by 21% while overall enrollment increased by only 2%.
 
A physical trading room is expensive to build and maintain. For example, a 2001 paper by Alexander, Heck, and McElreath (http://www2.stetson.edu/fsr/abstracts/vol_10_num1_p209.pdf) estimated the initial cost to be about $100k. A recent paper by Norton and Kim (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2048506) puts a lower bound of $100k, going up to $500k.  More recent estimates by Rise Vision puts the construction cost of a 30-seat trading room at around $500k and an annual budget of about $75k, and an additional $100k if you have a dedicated lab manager.
The answer to the question is not easy. It obviously depends on why you built the room in the first place and how it is used in the curriculum. 
 
1. If the trading room is like a library, just a repository of information and data, then the answer is simple: find a way for online students to have access, i.e. build a virtual trading room. Ironically, virtual access is the most difficult with some of the most costly resources in trading rooms, like Bloomberg terminals.
2. If it is integrated into a wide range of courses, then you have no choice but to create a virtual environment. By “integrated” I mean that professors regularly use the resources in their courses, for example to conduct trading simulations, assign projects utilizing the data resources, and so on. In this case, you also have to provide a way for students to learn to use the resources without necessarily having access to a person for help.
 
In our case, we have been working on creating usable technology for virtual trading rooms for almost two decades. When I was at CMU, we built a trading room in the early 1990’s, called the FAST Lab. It had live data feeds, data servers, etc. We had about 25 seats. We learned several things very quickly.
First, having the data, such as real time quotes, is not useful unless you developed educational tools to do something with it. This led to what is now the FTS Real Time System. Our goal here was, and remains, providing a set of tools so students can learn concepts, their practical application, and their usefulness. It is not “teach trading” or “stock picking.” Our teaching guide therefore contains a series of projects that students can conduct that help them understand risk and return, portfolio diversification, duration and bond immunization, how to use the option Greeks to hedge, how to apply valuation models, and so on. It provides a tight integration of theory and practice.
 
Second, we learned a lot about complexity. Putting a student into a trading room with all the “latest whatever” is like putting a drivers ed student into a formula 1 racing car and asking them to drive it at 200 mph. You have to step them up to that. Hence our effort on online textbooks, application guides, and so on. Our Financial Statement Analysis module is the latest in this series, and provides a structured, step-by-step way for students to learn financial statement analysis and conduct fundamental analysis. You can combine it with a trading exercise, but that is not necessary. As part of the module, we also provide quick and easy access to financial reports, so students don’t have to spend hours downloading data and copying and pasting; Excel skills are still needed, but for developing higher level skills, not the mechanics of data collection.
 
Third, we learned about access. Students tend not to work around a 9-5 schedule, and their needs depend critically on when something is due. So a room with 25 seats is totally inadequate if 300 students in core investments courses have a project that is due, and they all want to work on it at 1 am. So you have to provide distributed access. You also can’t do many other things with a limited number of seats. For example, the FTS Interactive Market is a trading simulation where students trade with each other. While the trading is organized around cases that reinforce conceptual understanding, students experience many things, such as market impact and liquidity, that are difficult to teach in a textbook and that cannot be experienced by simulations where they trade against the tape. If you want to use this in a class of 45 students, a 25 set room is limiting. So again, our application is accessible anywhere there is an Internet connection.
 
Anyway, we had a physical trading room for a few years, and then moved entirely to a virtual trading room. This also meant that when we expanded distance education and wanted to use these resources (for example, in the computational finance classroom in New York City), it was easy to expand. 
 
When writing this, I recalled an article published by the AACSB in 2003 that discussed may of the issues that we continue to discuss today. You can find a copy at http://gtf.mcmaster.ca/reports/p22-27.pdf    Many of the issues remain the same as what we discussed almost a decade ago.

Tuesday, August 7, 2012

Announcing the Financial Statement Analysis Module










We are excited to announce our new module: the FTS Financial Statement Analysis Module.    It is a unique hands-on and interactive way to teach/learn ratio analysis.  In this post, I’ll discuss why we developed this module, discuss some of the advantages of the approach we have taken, and how it complements Valuation Tutor.

In my mind, the main difficulty students have with financial statement analysis, and that instructors have with teaching it, is that there is a big gap between theory and practice.  Using a simplified “textbook” example abstracts away from all of the details you encounter with real-world financial statements.  Using the actual statements of only one or two companies is helpful, but the knowledge may not be easily transferable to other companies.  And the more you focus on actual statements, the more you focus on details and perhaps lose sight of the bigger picture.  And this can be boring.

So the challenge was to come up with something that keeps the focus on the conceptual framework and lets you see how the details you encounter in real world financial statements fit into this framework in an exciting way.  At the same time, we wanted a manageable way for students to gain experience and develop judgment by working with a large set of companies if they wanted to.

The FSA module achieves these goals.

First, there is emphasis on practice.  It uses only real world financial statements.  You can use the filings of any of the over 4000 companies we currently cover.  These are (mainly US) companies that file using the SEC’s interactive data format.   The filings are updated every night. Filings from other countries are not as easily accessible; we have a data project underway where we are manually converting statements of Australian, European, and Canadian companies into an accessible format.

Second, there is a conceptual framework through which you analyze companies. The topics here are fairly standard (profitability analysis, risk analysis, operations, earnings quality, and price ratios).  What is unique is that you literally drag the information from the statements and drop it into the conceptual framework.  This way, you see first hand what is needed for what analysis, how companies present information, what items you have to aggregate or disaggregate, and so on.

To help with this, we also provide a Self-Assessment data set.  In this data set, we have gone through the analysis for several companies.  The software then guides you step-by-step though the process of analyzing any company in the data set.  You can ask for detailed help or less detailed help, even no help.  You can practice this repeatedly.  For example, you can start with detailed help, and repeat the process with less help.  Or you can do the opposite: start with no help, and then turn it on if you get stuck.  The ability to practice really helps in bridging the gap between theory and practice.

For individuals, the FSA module provides an intuitive and interactive way to learn the theory and the practice.  The self-assessment data set guides you through the steps so you are not left guessing whether you are doing something correctly.  You can work at your own pace, repeating an exercise as many times as you like.  By saving your work, you can add to your knowledge over time without having to restart every time.

For instructors, we have additional tools, described under the “Instructors” section of this site.  You can assign you own practice data for the students for a company you choose.  You can also “grade” assignments by checking how student solutions compare to yours: you assign a problem and a company; the students work through the exercise and send you their answers; the software quickly compares your solution to those of all the students.

Two features we would like to highlight are:
--- the self-assessment mode, that lets students work though examples at their own pace
--- the instructor grading support, that provides an immediate evaluation of student submissions to your solution

We provide examples for students to work through, but these can easily be constructed and assigned by an instructor.  We also have a test bank dataset that you can use.

The motivation for creating this module came from Valuation Tutor (VT).  We received a lot of very positive feedback on Valuation Tutor and its ability to graphically compare companies along different dimensions.  But  VT assumes that students know how to construct ratios and other performance measures from financial statements, and so focuses on interpretation.

The feedback pointed out the need for a step-by-step way for students to learn how to construct ratios, essentially a pre-requisite for VT.  The FSA module fills exactly this gap.   Please look at the descriptions and videos at www.fsamodule.com and don’t hesitate to contact us for more information

Wednesday, April 4, 2012

Teaching Bond Valuation


I am struck by how different instructors use various parts of our system to teach essentially the same thing.   In fact, when someone asks how to teach something using our system, the answer mostly depends on your style of teaching.  This is hard to explain in the abstract, so let me give you an example. on teaching Bond Valuation.  A previous post shows you alternative ways to teach portfolio diversification.


The basics of teaching bond valuation include understanding discounting and the price-yield relationship, the term structure of interest rates, spot and forward rates, and perhaps managing interest rate risk using duration and convexity.   These topics are taught in at least 3 different ways:

Method 1: Bond Tutor - This is an interactive textbook; calculators are embedded into the online text, and students can experiment with concepts  as they learn about them.  For example, students can change the yield curve or forward curve, and observe the effect on the other curve and also on bond values:
 


Method 2: The FTS Interactive Markets-  Here, students trade bonds and learning takes place through price discovery and by building an analytical support system to help their trading.   In the simplest case (B01), there is a flat yield curve, so you learn how to discount given a yield.  Students who do not discount correctly lose money in the trading exercise to others.   Cases B02 and B03 introduce spot rates and forward rates.  Case B04 requires them to hedge a bond portfolio using duration.  Subsequent cases build on this, and the most advanced have trading of swaps as well as caps and floors to manage interest rate risk.  This is a synchronous “in class” exercise; students trade with each other, they react to each other.  Learning goes beyond a textbook; you need to understand the concepts, but you also learn how to use the concepts in making a trading decision.  The biggest way in which they lear is by building a decision support system in Excel.  The Student Case Preparation Manual shows you what students have to do to prepare for the case and how the learning takes place.  The following screen shows a real example of the FTS Interactive Trader with trading case B02:



Method 3: The Real Time System- The textbook and interactive markets are great teaching tools, but they do not let you experience some of the complexities of managing a real world bond portfolio.  The FTS Real Time system, with its built in analytics, provides some of this experience.  Here is a sample screen (of the Windows) of our Duration and Interest Rate Risk project




You can see my position in two bonds, and importantly, the portfolio analytics at the bottom right, and these analytics are used by students to make trading decisions.  

Summary These methods are not exclusive, and my description of what is available is not exhaustive (see below).   Bond Tutor represents the smallest deviation from a traditional textbook.  The FTS Interactive Markets are short trading exercises and it is easy to introduce the trading exercise in class and conduct a discussion that relates the concept to the trading exercise and also to markets and price discovery.  The FTS Real Time exercise abstracts away from price discovery but exposes students to application of the concepts to real-world interest rate movements.   

Just as an aside, we have several other modules for teaching fixed income; these include:

--The Treasury Calculator (the relationship between quoted Treasury prices and yields; yield curve approximations and interpolation; understand duration and convexity)

-- The Bond Immunization Lesson (a self contained interactive lesson for understanding how immunization works)

-- The Interest Rate Risk Module (Plot historical yields, spot rates, and forward rates, Animate curves to get a visual feel for long terms trends and mean reversion,  Calculate volatilities and correlations, Conduct principal component analysis, Backtest to understand the efficacy of different immunization strategies)

--The Principal Components Lesson (Understand the essence of principal component analysis with a visual calculator)

--The BDT (Black-Derman-Toy) Module (calibrate a BDT model to your own yield curve data using either yield volatilities or local vitalities, transfer the calculated lattice to Excel to price interest rate derivatives)





Friday, February 10, 2012

Experimental Research with FTS

Experimental research is a big part of our history.  In fact, the first version of the FTS Interactive Markets was developed in 1989 and 1990 for conducting experiments on information aggregation; the resulting paper came out in the Journal of Finance in 1991.  I even have a picture of the old text-based interface, on an original PS2 computer:
 
Since then, the FTS Interactive Markets have been used in many other experiments by different researchers.  Even today, we are helping several people with their new experiments; this includes setting up the trading parameters as well as software modifications to accommodate their needs. 
 
Beyond market experiments, though, we also have a “generic” experimental platform for conducting all sorts of experiments easily; these include behavioral experiments, auctions, and so on.   Examples of how to easily design eperiments are described in this document.

You can also use it to add non-market dimensions to a market; one example is adding a cheap-talk phase before a trading phase in a market, as in http://server1.tepper.cmu.edu/Seminars/docs/JobMarketPaper_0107_HongQu_CMU.pdf 
 
The experimental system is fairly straightforward. I’ll use an auction as an example.  You set up the experiment in an Excel workbook. One worksheet contains common information, such as instructions, sent to all subjects. Another worksheet contains information to be sent to individually, such as a private value or signal for the object being sold in an auction. This area also specifies what input is required from the subjects, such as a bid for the object being auctioned.
 
• You run the experimental server on your computer and connect it to your workbook.
• The subjects all connect to your computer using the client program.
• You send the common information
• You send the private information
• The subjects enter their responses (e.g. bids) during some period of time that you choose.
• When the time is up, you “grab” all their responses, calculate the outcomes in your
workbook (e.g. the price paid and who is allocated the object) and send this information
back to everyone.
 
The information flow described above can also be made continuous; in fact, you can mix and match
information that is updated constantly and that is updated manually. For example, if you are running a sealed bid auction, you only need to retrieve bids at the end of each round. If you are running an English auction, you may want to display the highest bid (or all bids) continuously.
 
The experimenter’s screen looks like this:
In the spreadsheet, you can set the colors of each cell, the text alignment, and you can see all the features in the screen shot.  We have a standard template that you can modify to create a variety of different types of experiments quickly and without a lot of effort.  If you have questions, please contact us, we love to talk about experimental research.

Monday, January 9, 2012

Portfolio Fundamentals


An exciting new capability is the integration of the Real Time System with Valuation Tutor (VT in what follows).  What it lets you do is conduct a fundamental analysis of your portfolio and also helps with stock selection.  I will describe the latter capability in a more general posting about using VT for stock selection.  Here, I will focus on the portfolio analysis.

Recall that in VT, you can analyze and compare companies along many dimensions.  These include common size analysis, cost-volume profit analysis, efficiency ratios, as well as price ratios.  Now, you can do the same for your stock portfolio.  This lets you answer questions such as: what is the ROE of my portfolio?  How does it compare to the stocks in my trading case?  How does it compare to specific industries or sectors?  And ROE is just one metric; you can compare your portfolio to any subset of stocks (in the VT dataset), sector and industry averages, and see exactly what are the fundamentals of your portfolio along any dimension covered by VT.

To use the capability, launch the FTS Real Time client and log in to one of our US Equity cases.  I will use the simplest case, the 30 stock case to illustrate the connection.  After you log in, click on the (new) Valuation Tutor tab:



 When you transfer the data, VT creates an (artificial) stock called “RT Portfolio”

Saturday, January 7, 2012

Valuation Tutor

We have had some exciting developments with Valuation Tutor. First, we have added a lot of visualization and comparison capabilities.

 

You can read about these on the Valuation Tutor blog:
Financial Statement Analysis- The Power of Visualization
Financial Statement Analysis- The Power to Compare 
The capabilities are also described in the video at the top of this blog. Second, we have integrated Valuation Tutor with the FTS Real Time Client. I will soon post a description of this, together with an example of using Valuation Tutor for stock selection, so stay tuned!