rstats-gsoc / gsoc2017 Goto Github PK

\section{Google Summer of Code}

\begin{frame}
\frametitle{Google Summer of Code (GSOC)}
Student gets $5500 for writing open source code for
3 months.
\begin{description}
\item[Feb] \textbf{Admins} for open source organizations e.g. R,
Bioconductor (MUGQIC?) apply to Google.
\item[Mar] \textbf{Mentors} suggest projects for each org.\
\textbf{Students} submit project proposals to Google.\
Google gives funding for $n$ students to an org.
\item[April] The top $n$ students get $500 and begin coding.
\item[July] Midterm evaluation, pass = $2250.
\item[Aug] Final evaluation, pass = $2750.
\item[November] Orgs get $500/student mentored.
\end{description}

I have participated since 2012 as an \textbf{admin} and
\textbf{mentor} for the R project.
\end{frame}

\begin{frame}
\frametitle{What makes a good GSOC project?}
Coding projects should:
\begin{itemize}
\item Result in free/open-source software.
\item Be 3 months of full time work for a student.
\item Include writing documentation and tests.
\item Not include original research.
\end{itemize}
Examples:
\begin{itemize}
\item Mathieu/François can be \textbf{admins} for MUGQIC org.
\item Robert/Dan can be \textbf{mentors} for a project to implement
new features in Gemini.
\item Any undergrad/master/PhD candidates (at McGill or not)
can be \textbf{students}.
\end{itemize}

\end{frame}

The first link on the GSOC page for R shows the 2016 projects. Should that link go to the projects list in the 2017 wiki?

Hi,
We, Dr.Veysel Kula and Dr.Ender Baykut, are working on a paper utilizing two-regimes MSGARCH model. In fact, this is our first attempt to use the model. Unfortunately, there is none to help us. We finally have had the attached output. But we are not sure whether we are on the right track. We will be pleased if you provide your comments on the results.

summary(XBANK_msgarch)
[1] "Specification Type: Markov-Switching"
[1] "Specification Name: sGARCH_normal_sym sGARCH_normal_sym"
[1] "Number of parameters in each variance model: 3 3"
[1] "Number of parameters in each distribution: 0 0"
[1] "Default parameters:"
alpha0_1 alpha1_1 beta_1 alpha0_2 alpha1_2 beta_2 P P
[1,] 0.1 0.1 0.8 0.1 0.1 0.8 0.5 0.5
[1] "DEoptim initialization: FALSE"
[1] "Fitted Parameters:"
alpha0_1 alpha1_1 beta_1 alpha0_2 alpha1_2 beta_2
[1,] 0.006208474 0.02679562 0.9731044 0.6692769 0.1155627 0.8843373
P P
0.974796 0.6499313

[1] "Transition matrix:"
t = 1 t = 2
t + 1 = 1 0.97479621 0.6499313
t + 1 = 2 0.02520379 0.3500687

[1] "Stable probabilities:"
Stable probabilities
State 1 0.96266853
State 2 0.03733147

[1] "Unconditional volatility:"
State 1 State 2
7.879386 81.8089

Log-kernel: -46533.55
AIC: 93030.33
BIC: 93082.4>

1) There are two regimes; regime 1 with low volatility, regime 2 with high volatility.
2) As beta 1 is higher than beta 2, regime 1 is more persistent.
3) Are both regimes significant?