ivo welch: 2021 MFE 404 syllabus

Introduction

The syllabus space website name is Site mfe404-welch-win2021 on syllabus.space. The passcode is the first name of an individual, which I shared with you in an email in December.

Meeting Times

Ordinary Class Meetings: Jan 5 - Mar 9, 2021.

Zoom, 08:30 to 11:15.

Final Exam: TBA

Basics

Mikhail Chernov's 400 course has covered basic financial markets, present values, risk and return, utility and risk aversion, portfolio choice, the CAPM, and multi-factor models. Although we will spend the first week recapping Mikhail's class subjects (and getting started on WRDS), this is just a quick review. You are already expected to know his earlier material. All of his course content is also fair game for my exam.

Before the first day of class, i.e., over winter break, please review Chapters 1-10 of the textbook, ed4.

Finance 404 is the second finance course in our 1.5-course sequence. 404 focuses less on the mathy nitty-gritty (which is important but covered in other courses) and more on the conceptual underpinnings of financial economics. This means that 404 is more intuitive than most of our more quantitative MFE courses---but it is not easier. It is different. (This MFE class also has some overlap with the Anderson Finance MBA corporate finance course, but it covers the subjects in a very different order and with a very different emphasis.)

Objectives

The point of this course and my textbook (we will be using the 4th edition!) is not to teach institutional details (which change year to year), but to teach the basics, methods, jargon, and ways of thinking of finance. You should learn to think of finance and economics as "common sense translated into algebraic formulas." A formula or recipe that you do not understand is useless. Always remain skeptical. If something appears not to make sense—here, in other classes, or in the real world—research and ask!

Knowledge of this material is truly the common ground upon which everything else in finance rests. All the advanced and institutional finance knowledge in the world will not help you if you have not mastered the material from this course cold. If you are solid in its building blocks, then you will succeed. This is regardless of whether you will work in an investment bank, hedge fund, consulting firm, academic economics or finance, or management.

Prerequisites

• The subject of finance is by nature technical. You need a thorough and comfortable ability to handle algebra and arithmetic. There are a lot of formulas and equations, but there is no higher math in this course. You need math aptitude, not math knowledge.
• You need to be able to pick up an accounting and a statistics introductory textbook, for reading up essential knowledge if you happened to have missed or forgotten some.
• You need good computing skills. This is not 1975. It is 2020 (a memorable year!).

This class will necessarily be easy for some students and difficult for other students. If you find that the class progress is perfect for you in terms of speed, you will be in the minority. If you find the material very difficult, you must take the initiative to keep up. The TAs and I are committed to provide as much help as we can. Yet, if you wait until the week before the midterm, then it will likely be too late.

Idiosyncracies and Teaching Philosophy

The single biggest difference between my course and similar finance courses elsewhere in other universities is that in my course, you will learn what really works in the real world and what does not. You will not learn what I wish would work.

In other classes, the models will seem to work and fit together nicely. The world will seem a happy place. The formulas will be nice recipes, ready to be applied. You may not even learn what they mean. You will feel that you have learned the recipes of finance. Unfortunately, it will be all wrong. These courses teach bed-time fairy tales, whose only function is to make you sleep better. Watch the basic explanation by Eric Falkenstein in Asset Pricing Theory Explained.

An analogy: if ninety percent of employers were creationist, would you want me to teach you the ins and outs of creationism? Or, would you want me to teach you first what creationists believe and then why it is wrong? Or, is teaching only creationism the "practical" way? I do have to teach you what the creationists believe, because you will confront their dogma all the time. But I will also teach you why and where they are wrong.

If you are looking for the simple "right" answers, I cannot give them to you. No one can. The real-world data is messier and more confusing than we wish it were. I think it will make you a better decision-maker to understand where you tread on safe grounds and where you do not. If you don't want to hear the truth, if you want to just learn how to apply formulas, take another course.

I also have other views worth noting, both about finance and about teaching finance. I already described them above, but it's worth repeating:

• I care primarily that you understand the basics of finance, inside and out. I care that you understand why you are using them. I care that you understand the limitations of what you know. I care that you can apply what you know.
• I do not want you to memorize twiggles. I believe that most of you will forget such details. One year after the class, you wouldn't remember them. I tend to forget them, too. I want you to understand the "system."
• I will want you to do well in interviews. This requires solid knowledge of the basics and an ability to apply them quickly on questions and scenarios (which is what we will be doing a lot). Interviewers rarely ask about institutional details, and they don't do complex 30-page case studies, either. (If nothing else, the interviewers don't want to argue about the less relevant and/or ambiguous aspects about cases in job interviews, either.)

Mutual Responsibilities

Despite rumors to the contrary, a business school is not a service organization. UCLA Anderson is not Taco Bell, and we don't sell degrees. Do not view yourself as our customer. If anything, our customers are our alumns and recruiting employers. Our reputation rests more with them than with you. You are "only" the product, and we don't want defectives.

My class is difficult. You can watch fun Ted talks or easy PBS shows yourself without me, as can anyone who is not attending business school. I am here to convey a rigorous and difficult subject matter, not to entertain you. Do not confuse me for an employee, for customer service, or for tech support. I am a professor of finance.

Your education is your job. I am only here to help you if you want to learn the material. Again, the material is difficult. I don't deliver knowledge via silicon implants. My job is to help you if you want to learn, and do so in any reasonable way I can. I take this very seriously. My TAs and I are here to help. Talk to the TAs and/or me me when you have problems. Make use of the TAs' perspective. If you worry that you are one of the weaker students, I will personally (from my budget) pay to get you extra 1-on-1 tutoring help with a TA. Talk to me. There will be no excuses that you did not have the help you needed. Where there is a will, there is a way (and we will pave it for you but not walk it for you).

So, again, it is entirely your responsibility to learn. I can only aid you in your endeavor to learn. I want to help you learn, to the best of my own ability. This is why I want to be a finance professor in the first place—to help students who really do want to learn to actually learn. I could care less about students who think it is my job to entertain them or to make them understand. It is the students' obligation to lean the material, not mine. I already know the material. And I do not repeat my classes in private, e.g., because you were busy with other things.

In the end, it is up to you what you make of this course. You should try your best to stay on top of the current material, and take advantage of all resources offered to you. If you find yourself seriously behind, consider hiring a tutor—the earlier the better. Beginning one week before the exam may be too late. If you cannot afford hiring a tutor, talk to me. I have contributed to paying for tutors for students who want to learn and who could not afford one. Again, my goal is to help, in any way possible, those who want and deserve to be helped. I love to teach such students. It's what I want to do in my life.

Workload Warning

Do not assume this class is easy or works on cruise-control. The suggested out-of-course workload is targeted for an average of 6 hours per week, sometimes more, sometimes less. With class time, this is 10-12 hours/week.

Be religious in your course preparation: print and read the classnotes, do the homework, read the book chapters, attend class. Don't expect to be spoon-fed. Be prepared to research novel problems by yourself. Tackling novel problems is an important skill.

Readings

Classnotes

The class notes are the definite source of required material and the basis of 90% of the exams. They are available on the book website. Although the class notes are posted, I may change them as time goes by. I always try to improve them. Please print the slides about 1 week before they are used in class. (Otherwise, you may have to look at a slightly different set of notes from what will be used in class). Class notes are not spaced by class, but spaced by topic. We should be able to cover roughly 10-25 pages of notes per class.

Expect to be randomly called upon during class to answer questions in the class notes.

Textbook

The textbook for this course is Ivo Welch, Corporate Finance: An Introduction, 4th ed (MFE). Although the book is free online, I would suggest that you purchase a printed copy. This way, you can mark up text and you are immune to website downtime. (My website may well go down for a week before the exam; I am not a pro sysadmin.) The book is deliberately priced very cheaply, at cost. Other books cost 4-5 times as much.

Ungraded Practice Quizzes

You can also test yourself on the syllabus.space course website. There are questions for each book chapter.

Beyond

There are many other useful sources of information. For example:

• You may also find it useful to have access to a book on valuation, such as Koller, Goedhart, Wessels Valuation (McKinsey Press).
• I can also recommend some other corporate finance textbooks: Berk-DeMarzo Corporate Finance, Allen-Brealey-Myers' Fundamental of Corporate Finance; Ross-Westerfield-Jaffe Corporate Finance; and the higher-level Grinblatt and Titman Financial Markets and Corporate Strategy. The GT text is more complementary to my own textbook than the first three.
• Professional Readings: Everyone in the class should regularly read business publications such as The Wall Street Journal or The Economist. You absolutely cannot walk into an interviewing situation without knowledge of (and/or an opinion about) current events. Similarly, you should not walk into a financial services or consulting interviewing situation if you are not tech-savvy. For example, what are the best financial websites and why? Where can you get financial data on the web?
• Professional speaking and writing is very important. Take an acting course. It helps you to learn how to present yourself. This is especially useful for foreign students.
• The ability to manipulate data quickly is very important. For MBA students, take a computer programming course. For MFE, this is of course bread and butter. The world is becoming more analytical and data-oriented. Are you sure you are equipped to handle it?

In Class Setup

I want you to learn. I will always welcome challenge(s). I am not (too) sadistic. I am never malicious. I am not (too) arrogant, and I can take it as well as I can give it. I am hard to offend. Please participate. Enjoy this class. Work is serious enough. Make the class fun!

I am a believer in the Socratic method and learning by doing. I am a fan of Prof. Kingsfield.

I have a very strange sense of humor. Please don't take offense to inappropriate remarks. I try to control myself, but I am not always politically correct. If you are very sensitive, and you support the UC Microaggressions list, then please let me know ahead of time. (If there are alternative offerings, enroll there if you can. Unfortunately, 404 has no alternative instructors for now.)

Necessary Class Preparation

• Read the relevant book chapter before class. You will not understand everything, but you should understand the "drift." Having preread the material should help you to follow the discussion in class—and enjoy the cold-calling more.
• Print and look over the relevant classnote(s) ahead of time.
• Attend Zoom class. Have an ordinary calculator at hand. Have the classnotes for the previous, the current, and the next session.
• Read the relevant classnotes and book chapter again after class, and in more detail.
• Do the assignments and homeworks.

Attendance

You are not required to attend. However, if you do attend, be prepared to be called upon.

This is my first (and hopefully last) time teaching over Zoom. In real classes, I encourage questions and interruptions. Here, it may be more difficult. Let's see how it goes. I am a private pilot, and ATC follows an efficient protocol to avoid "stepping" on other communications. Let's try something like this in my class:

• If you want to question or say something out of order, start by saying "Prof Welch, <your name&rt;" and wait for me to acknowledge you by calling your name. If I do not respond, try again. (If multiple students want to talk, wait until the conversation is finished, and then talk.)
• End with "thanks" to help me know when to start answering.

I have also asked the MFE office to arrange for recording and review of sessions. Please do not share these videos with anyone.

Grading

Overall Component Scoring

Experience tells me that most of the grading variation will come from midterm and final. Only 1 in 10 students is affected by all the other parts. A good rule of thumb is that everything not-exam related matters only for students that are close to a letter-grade margin.

Letter grades are relative to other students. The class roughly obeys the quasi-mandatory Anderson grade curve—about 40% A's (15% A+), 45% B's, 10% C's, and 5% F's, although I have rarely upped this on occasion when it was clear that my class consisted of more than a fair share of great students. (As far as your transcript is concerned, you should neither be punished nor rewarded for taking this course. I am not here to fight or condone grade inflation or deflation.) The TAs and I will give it our best efforts to be appropriate and fair, but grading is never perfect. There is inevitable noise.

In assigning course letter grades, I often put more weight on the better of the midterm and the final. After the midterm+final computation, I would add in the project score, the homework score, and my discretionary score. If you are not present for the midterm for a good reason, we will use your final as the sole grade determinant. This incurs a very modest penalty, though, in that it means that (a) you are losing the information of where you stand and (b) the option of overweighting the higher of the two. But you can get an A+ without taking the midterm (good excuse) and without doing any homeworks or projects—it's just a little harder.

Exam Format and Preparation

There are no surprises in my exams. There never have been. They are just like my lectures, which are just like my slides, which are just like my book. One student last year told me that my lectures were not adding a lot over the book, and he was perfectly fine missing some---he ended up getting the 5th-lowest score on the final. The material is deceptive: it is a lot harder than it seems. Yes, you may think the material is somewhat redundant, but it takes redundancy to absorb it. And this is exactly the material that you need to know inside-out to succeed in finance. If the material seems too easy for you to bother with, you are putting yourself in danger of a low grade.

Many of my old exams are on my course website.

The midterm questions are basically the same format and level of difficulty as the homeworks. If you have no problems answering similar questions under time pressure, you will also have no problem in the midterms.

The final is similar to the midterm in format, but cumulative in coverage. If you cannot make the scheduled final on (with prior justification sent to the TA team), then you can attend the makeup final. If you miss both final exam dates, then you have the choice between taking an oral exam with me that I will administer (my advice: avoid this!) and taking the final exam next year from another instructor teaching the same course.

Regrading Policy

To the extent that grading is relative, we do not want to make this a race of nitpicking. You learn nothing by squeezing the last point out of your obscure answer to question 12. We try to be (as) accurate (as possible) in grading, and we do not take it lightly. However, it is not possible to get everything perfectly right. Every student will get a few extra points on some questions where they don't deserve it, and a few missed points where they did not deserve it. It should roughly average out. The average question grade is more precise than an individual question grade. It would be unfair to those never asking for a regrade not to take this into account. Of course, if there is a big mistake in grading, we want to know. Our most important job here, just after providing you with the knowledge you need, is to keep the grading fair.

My discretion rarely makes a difference. I have never pushed a student more than one letter notch up or down and only if they started on the correct half. That is, if a student stands at 3.1, I can push to a B+, but not to a B-. Similarly, I can push a 3.6 from an A- to a B+.

We regrade only full exams, not just specific questions, except in cases where we obviously blundered (such as missing that you wrote something at all).

The TAs will make small mistakes grading all exams, some in the student's favor, some against. Because grading is relative to other students, we do not want to systematically reward nitpicking. However, we also do not wish to make a systematic mistake in particularly hard cases. So, our regrading policy is as follows.

A student requesting a regrade can choose between one of two options:

1. A regrade only of specific questions in dispute. However, this "costs" an immediate penalty of one question. Thus, a student with two incorrectly graded questions in favor and two against should choose this option.
2. A regrade of the entire exam, with a clear directive for the TAs to reexamine every answer. Thus, a student with one incorrectly graded question in favor and none against should choose this option. It is possible for the overall exam score to decline when this option is chosen.

We do make random copies of exams. We have caught at least one student in the past who thought it was a good idea to clarify his answers in the exam with the same pen after the fact. We disagreed, even though his answers were indeed greatly improved. His grade was not. He was duly expelled.

Book Homeworks

Book homeworks are really more or less optional. If you do them, you get a tiny bonus that will matter only if you are one of the two students nearest to a letter-grade cut-off. Otherwise, they do not matter.

The homeworks are the questions at the end of each chapter of the book. They are due seven days after completing a chapter, and even if the receptacle upload date is open longer than this. (We usually leave it open another day for stragglers.) Of course, you can work ahead on homeworks if you so wish. Clinton will coordinate homework collection and keep the grading sheet.

Computer Homeworks and Project

Honor Code

UCLA Anderson has an honor code, which prohibits cheating and plagiarism. Please report any violations to me.

You must agree not to cheat on the exams; and to inform me if you see other students cheating. Seating space may be at a premium, and it is tempting to glance at your neighbor's solutions. Do not do it: the penalties far outweigh the benefit. If I catch a student, I will try to press for the maximum possible penalty—and I have a reputation to defend: I have caught cheaters in the past.

We had an unpleasant experience a few years ago. Don't cheat. It is not worth it. We have expelled students for cheating. Please bring any cheating to my (not the TA's) attention.

Timeline

Every topic in our teaching schedule should take about 2 weeks, plus or minus 1 week. About 1 in 4 classroom sessions can be quite short in order to look jointly over your programming assignments. All programming assignments have to be your own work, i.e., they are not group assignments. Writing a simple computer program is also a fair-game request on the midterm or final exam.

I do not use powerpoint but pdf classnotes. You can scribble the answers on an ipad with a pencil; or you can print the relevant classnote(s) and scribble with a pencil. (PS: Transcribing your pencilled notes afterwards usually improves both your learning and the overall classroom atmosphere—even though answering your emails will have to wait until after class. Please abide by this policy without me having to remind you.)

Iff the server is down on task submission dates for more than 30 minutes within 2 hours of midnight, the homework will be due the following day at 5pm.

Again, the schedule is approximate. The last class is March 9.

Computer Assignments

The computer assignments are loosely connected to the rest of the course. They are useful, relevant, and necessary, but they are not 1-to-1 in sync with the material covered in the class itself.

One good and/or suggested computer language in this course is R. It is the standard in the industry. It sucks. R does not know whether it wants to be a parsimonious programming language or an interactive environment, so it has lots of inconsistencies. The worst R aspect is that errors are ridiculously easy to commit and incomprehensible to debug. Error messages don't even reliably give a line number. The best debugging advice: insert a lot of cat statements in your code and see where the error occurs.

Modern C++ based on the STL is also acceptable, but much harder. (Old-style object-oriented C++ is best forgotten.) Matlab is acceptable, but we have no expertise in it, either. Matlab is rapidly losing ground to R in the real world. Excel, Basic, etc., are not permissible for programming assignments in this MFE404 class.

Data work requires your own WRDS account. Most data assignments are based on CRSP data. Some are based on S&P Compustat data. (You are welcome to use the Bloomberg terminals in the Anderson library instead, but do not expect any help if you do.) You are expected to hand in written reports that are between 1 page and 5 pages long (the following week after I noted in the syllabus), and post your program on a website in case the TA wants to check.

Computer Homeworks

When a (CRSP) computer assignment requests results up to "now", it means up to whatever data WRDS already provides. This is usually up to the year before last year.

Your homeworks should be uploaded and then may be posted (publicly!) on the class website. Your future potential employers can see it! The submission format must be plain html or markdown, so your homework will be viewable in a browser. I don't mean a .docx file saved as an .html (or .md) file. I mean real clean html or markdown. Learn basic html and/o markdown if you don't know this yet: html, head, title, style, body, div, span, hr, p, table, tr, td, th, ul, ol, dl, li, dt, dd, h1, h2, h3. Learn basic css styles, but note that we will use a standard style sheet https://www.ivo-welch.info/teaching/hw.css to make every student's output look similar and non-gaudy.

Computer homework hand-ins have to follow a strict outline:

1. an executive summary page, with boldfaced key "take-away" numbers;
2. the homework text;
3. tables or figures (note that your code can create them!);
4. the computer code.

As noted, homeworks are only mildly related to the specific subject covered at the same time. Nevertheless, neither is redundant.

Please note that for large data pull requests from WRDS, you should download data once and then save your input data to a local file. This will make your and WRDS' lives (and debugging) a lot happier.

hw1 (indexes)

Learn WRDS!!

Indexes: For the (CRSP) S&P500 with dividends, what was the average arithmetic and geometric historical mean rate of return for daily returns, for monthly returns, for annual returns, and for 5-year returns from 1/1/1973 through 1/1/2015? Do the same for excess rates of return above the prevailing short rate, the 30-day Treasury rate, the 1-year Treasury rate, and the 5-year Treasury rates, respectively. This answers the questions of "how would each of these four self-financed strategies have performed?" (Does overlapping the longer-term series lead to different inference?)
Optional future assignment = how did HML, SMB, XMKT, etc. do in the last 5 / 10 / 20 / 50 years? What was the worst month?
Note: This assignment overlap with one in Bernard Herskovic's course.

hw2 (evtstudy)

We are interested to learn how stocks perform in terms of excess rates of return when they are actually paying a dividend. (The cum to ex-dates are noted for each stock in CRSP.) What happened to the stock return (minus the S&P500 on the same day) from days -20 through day +20, each day by day? Line up all stock excess returns on the event dates, and calculate the mean across all stock-days. This is the day 0 event mean excess rate of return. Eventually, your result will be 41 mean excess rates of return, that you can graph with relative day on the x-axis and mean rate of return (across all excess returns) on the y-axis. You can do this assignment only for calendar years after 2010 to make it simpler. (O: How does each stock's rate of return relate to the amount of the dividend it actually paid?).

hw3 (annretset)

Create an annual data set of compounded rates of return for each stock: perm, ticker, cusip, year, marketcap, rate of return. do not drop stocks in their year of disappearance! (This will be used later. For the hand-ins, just provide good summary statistics describing your data set. Do not try to submit the entire data set.)

hw4 (finlset)

Create an annual data set of earnings, cash flows, and leverage ratios (TL/TA [= total liabilities divided by total assets], market and book-based), as available to investors for sure in December of the year. Note that a firm that closes its financials in December 2012 usually won't release these financials until March 2013. Also, you may want to read ahead to learn more about what earnings mean if you do not remember your accounting courses. (This will be used later. For the hand-ins, just provide good summary statistics describing your data set. Do not try to submit the entire data set.)

hw5 (reconciliation)

How do you check your and others' results? There is really only one way: independent replication. (It is also helpful to trace individual stocks and to look at extreme values.) For our reconciliation application, return to your computer homework 2 and computer homework 3. Each student is paired with one other student (ideally, different students for both homeworks). Each student needs to share his/her writeup, take the other's data set, merge it with their own, and reconcile what is different:

• For homework-2, create a PERM,YYYYMMDD,RET0 (rate of return at event day 0) data set for a merge with the other's data set. Each student, take the other's data set, merge it in, and diagnose the individual data points that are different.
• For homework-3, do the same for all PERMNO's that begin with 100, i.e., 100*.

As the handin, write a report of what was different, how you would reconcile it, and what, if anything, you learned about your and their code from this exercise. The point is not to trash your costudent, but to learn therefrom.

hw6 (comps)

P/E: Produce a figure with E/P on the y-axis, and (the 3-year average of the) lagged change in percent earnings on the x-axis, one dot for each stock. (Optionally, size the dots by log-firm-cap.) Truncate both X and Y at the 10th and 90th quantiles.

Extra Credit: Scrape Yahoo!Finance to do it in real-time. You can use an extra 5 pages for this.
Note: In this assignment, it may help to look back on David Aboody's course.

hw7 (fm-mcbeth)

Do a basic Fama-Macbeth type of regression predicting cross-sectional stock returns with independent variables: firm-marketcap, price-normalized accruals, the earnings-price ratio, and 1/price. You can run this based on the annual data sets you created earlier. (If annual, you want to predict year t calendar stock returns with year t-2 variables, because a firm could end its fiscal year in May and report results in October.) The minimum requirement is Fama-Macbeth overall coefficients from one predictive set of (four) independent variables. (Each Fama-Macbeth coefficient is the average over the cross-sectional coefficients.) Extra Credit: are the differences between being scrupulous in knowing the variables (e.g., a 6-month lag post earnings disclosure), and one where you are "casual" (i.e., wrong)? Extra Credit: Any other variables that predict future calendar year returns reliably?
Note: This assignment has modest overlap with one in Lars Lochstoer's course.

Recognize that you have to make choices in how you deal with missing data, strange date, etc. Spell out and justify your choices. When do they matter?

Other Issues

Conflicts and Problems

If you have problems with me or the TAs, our strong preference is that you please come to see us first—either the person who you have problems with, or one of the other two members of our teaching team. (You can ask to keep your identity confidential, and we will honor this.) We want to make this class a great experience for everyone involved, and we are open to learning from you how to do it better. (In particular, I know that I occasionally go over the top with my brand of cynical and politically incorrect sense of humor, even though I wish I did not.

Electronic Attendance

Each student is required to check the class web site at least once every day. Important news will be flagged at the top. Students are responsible to have read posted announcements. They may even contain new assignments, and it is not an excuse that you have not bothered checking the website. Similarly, my computer died is not an excuse—backup is your friend.

The main computer system is https://syllabus.space. Expect some glitches. First, register in the system, and then enroll into course named welch-mfe404-2021.ucla.

Office Hours and Learning

We have deliberately scheduled TAs for two different days of the week. Note that I (the instructor) have already tried to explain the answer to questions in my book, lectures, and classnotes. Thus, I believe it may well be better if you first see if you can get another perspective from another person. This may well be a co-student. It may be a TA to give you a different perspective. (Being confused at first when learning is normal! Your goal should be to unconfuse yourself by trying yourself and by asking questions.) The TAs will also try to answer questions by email. Of course, if none of this works, please come and see me. I am here to help (though not usually by email).

I become very grumpy when students do not attend, have not prepared well, and then ask for private class sessions from TAs or me. I become understanding when a student could not come to class due to an interview, but then spent many hours trying to catch up, and comes to see costudents, TAs, and me for help for something that really did not click.

Do not expect the TAs to know everything. They are there to try to help you, but they are not me—and, besides, even I do not know everything about finance, either!