The main objective of this course is to develop the skills needed to do work in the industry or research with financial data. The course aims to provide students with techniques and receipts for estimation and assessment of quality of financial models. Each student is expected to choose a project related to the field of financial econometrics, and to make a report and a final presentation to the class. In addition, the student is expected to discuss about the advancement of the project at least once during the semester. The presentation can include theory, numerical simulations and/or data analysis.
The evaluation will be based on the report (50% of the final grade) and the final presentation (50% of the final grade). The report is expected to be typed in Latex, and should be around 15-20 page long. It should look as much as a reasearch paper as possible, with different sections and including a bibliography. The presentation will be 15 minute long, including 5 minutes for questions. Tentatively, this presentation will be held in a room on Keio campus (but it may also be conducted over Skype). You can use Overleaf where you can code directly online (and even share it with a friend). Also, you will have information for documentation and/or installing Latex at The Latex Project. There is no need to install Latex if you use Overleaf, so this is the simplest way to use it.
Students are required to bring their personal laptop to each lecture. They will be autonomous (in class) by following and implementing the methods from Introductory Time Series with R from Paul S.P. Cowpertwait and Andrew V. Metcalfe as a workhorse for their project. They are highly encouraged to ask questions about anything (even if it feels stupid).
Dataset can be found here: Maine.dat ; USunemp.dat ; cbe.dat ; pounds_nz.dat ; global.dat ; Herald.dat ; wave.dat ; Fontdsdt.dat ; ApprovActiv.dat ; motororg.dat ; wine.dat. To import the data on R, you can follow the example on specialresearch.R using the dataset file. You can also use other datasets here .
This class is not all about R, yet (some) learning of this programming language is key to understand thoroughly the material of the book.
Students need to be at ease with basic grammar of the language so that they can quickly interpret the output obtained when following the book content.
I do know that some students have NO background at all in programming language, and it requires some kind of effort as the notion used in the book can be sometimes
a little bit tough for a student with no notion on R (or any other programming language. This is a reason why I give more details about R in this section.
Again, the goal of this class is not that you become an EXPERT of R, but rather that you know enough so that you can read through the book content without being
completely lost. Consider your knowledge on R as your lifebuoy, so that you should make a wise investment on how many hours you wanna put it to learn the basics of it.
First, you need to download R (without R, you cannot follow the book). You can actually download R
here. Now that you have installed it, you can start to launch it, and you will see the R console.
Try to type 2+3, and then press "enter", and you should see 5 as a result. In principle, you can use only R console for this class, but this is not the easiest way.
In particular, each time you will use the console, you need to type again your code, or copy-paste from an annex file. This is rather time-consuming and you will
make more errors doing this, so I require students to also use the friendly interface user for R called Rstudio. RStudio is available
here. Once you have installed RStudio, you can launch it. From now on, you do not need to use the
R console, so you do not need to launch R application directly again (except for the part 1.2 of the tutorial on R that follows). Now it is time that you work
go through Chapter 1 and Chapter 2 of this tutorial.
Students can choose by themselves a project in any aera closely related or not to volatility estimation. Note that this is my area of expertise (and also the title of this class), so of course I encourage students to choose a project related to volatility estimation. For those who do not have any background or any specific idea for their projects, I provide those two types of models to choose time series. You can also find the original paper on ARCH model from Robert Engle. Finally, you can also get the paper on GARCH model from Tim Bollerslev. Note that you need to be on Keio campus to download those papers for free. Those are highly advanced Ph.D. level course notes and research papers, and this is expected that students are completely lost at first glance.