Changeset 639


Ignore:
Timestamp:
Sep 6, 2006, 2:18:46 PM (15 years ago)
Author:
Markus Ringnér
Message:

Fixed documentation bugs

Location:
trunk
Files:
2 edited

Legend:

Unmodified
Added
Removed
  • trunk/c++_tools/statistics/LinearWeighted.h

    r616 r639  
    5757    inline double beta_err(void) const { return sqrt(beta_var_); }
    5858   
    59     ///
    60     /// This function computes the best-fit linear regression
    61     /// coefficients \f$ (\alpha, \beta)\f$ of the model \f$ y =
    62     /// \alpha + \beta (x-m_x) \f$ from vectors \a x and \a y, by
    63     /// minimizing \f$ \sum{w_i(y_i - \alpha - \beta (x-m_x))^2} \f$,
    64     /// where \f$ m_x \f$ is the weighted average. By construction \f$
    65     /// \alpha \f$ and \f$ \beta \f$ are independent.
    66     ///
     59    /**
     60      This function computes the best-fit linear regression
     61      coefficients \f$ (\alpha, \beta)\f$ of the model \f$ y =
     62      \alpha + \beta (x-m_x) \f$ from vectors \a x and \a y, by
     63      minimizing \f$ \sum{w_i(y_i - \alpha - \beta (x-m_x))^2} \f$,
     64      where \f$ m_x \f$ is the weighted average. By construction \f$
     65      \alpha \f$ and \f$ \beta \f$ are independent.
     66    **/
    6767    void fit(const utility::vector& x, const utility::vector& y,
    6868             const utility::vector& w);
    6969   
    7070    ///
    71     /// Function predicting value using the linear model: \f$ y =
    72     /// \alpha + \beta (x - m)
     71    ///  Function predicting value using the linear model:
     72    /// \f$ y =\alpha + \beta (x - m) \f$
     73    ///
    7374    double predict(const double x) const { return alpha_ + beta_ * (x-m_x_); }
    7475
     
    8081    { return sqrt(alpha_var_ + beta_var_*(x-m_x_)*(x-m_x_)+s2_/w); }
    8182
    82     ///
    83     /// estimated error @a y_err \f$ y_err = \sqrt{ Var(\alpha) +
    84     /// Var(\beta)*(x-m)^2 }.
    85     ///
     83    /**
     84       estimated error \f$ y_{err} = \sqrt{ Var(\alpha) +
     85       Var(\beta)*(x-m)} \f$.
     86    **/
    8687    inline double standard_error(const double x) const
    8788    { return sqrt(alpha_var_ + beta_var_*(x-m_x_)*(x-m_x_) ); }
  • trunk/doc/Statistics.tex

    r627 r639  
    276276The polynomial kernel of degree $N$ is defined as $(1+<x,y>)^N$, where
    277277$<x,y>$ is the linear kernel (usual scalar product). For the weighted
    278 case we define the linear kernel to be $<x,y>=\sum w_xw_yxy}$ and the
     278case we define the linear kernel to be $<x,y>=\sum {w_xw_yxy}$ and the
    279279polynomial kernel can be calculated as before
    280280$(1+<x,y>)^N$. Is this kernel a proper kernel (always being semi
Note: See TracChangeset for help on using the changeset viewer.