BMH_layout.txt

# Bayesian Methods for Hackers Layout

\section{ Preamble}


\section{ Introduction }
    \
    \subsection{How does bayesian inference differ?}
    \subsection{ PyMC }

        
\section{Further PyMC}
    #flexible about what this section is. Basically it's more intro to the 
    syntax of PyMC, with examples + distributions.

\section{ Would you rather lose an arm or a leg? }
  #Introduction to loss functions and point estimation.
  \subsection{ Loss functions for parameters}
  \subsection{ Why we should be interested in expected values }
     \subsubsection{ Optimization with scipy.optimize }
  \subsection{ Point estimate for predictions }

\section{ The greatest theorem never told }
    #This is about the law of large numbers and how a bayesian uses it for estimates. 

\section{What should my prior look like?}
    \subsection{Noninformative priors...}
    \subsection{Noninformative priors do not exist}
    \subsection{Good choices of priors }
  
\section{ Bayesian Networks }
    #I do not know too much about this.

    
\section{More hacking with PyMC}
    #some examples from that PyMC website.


\section{ More examples of Bayesian analysis}
    
    \subsection{ For the interested: We already do perform Bayesian inference! }
        #This is about least-squares solution equivilance to minimization of normal errors, similarly Lasso and Elastic Net solutions.

    
    
\section{ Conclusion }


\section{Appendix}
    \subsection{A}
        #Chart of distributions and their support
    \subsection{B}
        #Appendix on MCMC
    \section{C} 
        #Proofs