class RooTreeData: public RooAbsData

   Interface function to TTree::Scan

 Interface function to TTree::Scan

 Interface function to TTree::GetEntries

 Interface function to TTree::Reset

 Interface function to TTree::Print

 Interface function to TTree::Fill

 Interface function to TTree::GetEntry

 Interface function to TTree::Draw()

 Interface function to TTree::Draw()

 Create TTree object that lives in memory, independent of current
 location of gDirectory

 Destructor

 One-time initialization common to all constructor forms.  Attach
 variables of internal ArgSet to the corresponding TTree branches

 Initialize cache of dataset: attach variables of cache ArgSet
 to the corresponding TTree branches

 Load TTree name 'treename' from file 'filename' and pass call
 to loadValue(TTree*,...)

 Load values from dataset 't' into this data collection, optionally
 selecting events using 'select' RooFormulaVar

 Load values from tree 't' into this data collection, optionally
 selecting events using 'select' RooFormulaVar

 The source tree 't' is first clone as not disturb its branch
 structure when retrieving information from it.

 DEBUG: Dump contents

 Remove tree with values of cached observables
 and clear list of cached observables

 Change name of internal observable named 'from' into 'to'

 Activate or deactivate the branch status of the TTree branch associated
 with the given set of dataset observables

 Cache given RooAbsArgs with this tree: The tree is
 given direct write access of the args internal cache
 the args values is pre-calculated for all data points
 in this data collection. Upon a get() call, the
 internal cache of 'newVar' will be loaded with the
 precalculated value and it's dirty flag will be cleared.

 Load the n-th data point (n='index') in memory
 and return a pointer to the internal RooArgSet
 holding its coordinates.

 Add a new column to the data set which holds the pre-calculated values
 of 'newVar'. This operation is only meaningful if 'newVar' is a derived
 value.

 The return value points to the added element holding 'newVar's value
 in the data collection. The element is always the corresponding fundamental
 type of 'newVar' (e.g. a RooRealVar if 'newVar' is a RooFormulaVar)

 Note: This function is explicitly NOT intended as a speed optimization
       opportunity for the user. Components of complex PDFs that can be
       precalculated with the dataset are automatically identified as such
       and will be precalculated when fitting to a dataset

       By forcibly precalculating functions with non-trivial Jacobians,
       or functions of multiple variables occurring in the data set,
       using addColumn(), you may alter the outcome of the fit.

       Only in cases where such a modification of fit behaviour is intentional,
       this function should be used.

 Utility function to add multiple columns in one call
 See addColumn() for details

 Split dataset into subsets based on states of given splitCat in this dataset.
 A TList of RooDataSets is returned in which each RooDataSet is named
 after the state name of splitCat of which it contains the dataset subset.
 The observables splitCat itself is no longer present in the sub datasets.
 This method only creates datasets for states which have at least one entry
 The caller takes ownership of the returned list and its contents

 Plot dataset on specified frame. By default an unbinned dataset will use the default binning of
 the target frame. A binned dataset will by default retain its intrinsic binning.

 The following optional named arguments can be used to modify the default behavior

 Data representation options

 Asymmetry(const RooCategory& c) -- Show the asymmetry of the data in given two-state category [F(+)-F(-)] / [F(+)+F(-)].
                                    Category must have two states with indices -1 and +1 or three states with indeces -1,0 and +1.
 Efficiency(const RooCategory& c)-- Show the efficiency F(acc)/[F(acc)+F(rej)]. Category must have two states with indices 0 and 1
 ErrorType(RooAbsData::EType)    -- Select the type of error drawn: Poisson (default) draws asymmetric Poisson
                                    confidence intervals. SumW2 draws symmetric sum-of-weights error
 Binning(double xlo, double xhi, -- Use specified binning to draw dataset
                      int nbins)
 Binning(const RooAbsBinning&)   -- Use specified binning to draw dataset
 Binning(const char* name)       -- Use binning with specified name to draw dataset
 RefreshNorm(Bool_t flag)        -- Force refreshing for PDF normalization information in frame.
                                    If set, any subsequent PDF will normalize to this dataset, even if it is
                                    not the first one added to the frame. By default only the 1st dataset
                                    added to a frame will update the normalization information

 Histogram drawing options

 DrawOption(const char* opt)     -- Select ROOT draw option for resulting TGraph object
 LineStyle(Int_t style)          -- Select line style by ROOT line style code, default is solid
 LineColor(Int_t color)          -- Select line color by ROOT color code, default is black
 LineWidth(Int_t width)          -- Select line with in pixels, default is 3
 MarkerStyle(Int_t style)        -- Select the ROOT marker style, default is 21
 MarkerColor(Int_t color)        -- Select the ROOT marker color, default is black
 MarkerSize(Double_t size)       -- Select the ROOT marker size
 FillStyle(Int_t style)          -- Select fill style, default is filled.
 FillColor(Int_t color)          -- Select fill color by ROOT color code
 XErrorSize(Double_t frac)       -- Select size of X error bar as fraction of the bin width, default is 1


 Misc. other options

 Name(const chat* name)          -- Give curve specified name in frame. Useful if curve is to be referenced later
 Invisble(Bool_t flag)           -- Add curve to frame, but do not display. Useful in combination AddTo()
 AddTo(const char* name,         -- Add constructed histogram to already existing histogram with given name and relative weight factors
 double_t wgtSelf, double_t wgtOther)

 Create and fill a histogram of the frame's variable and append it to the frame.
 The frame variable must be one of the data sets dimensions.

 The plot range and the number of plot bins is determined by the parameters
 of the plot variable of the frame (RooAbsReal::setPlotRange(), RooAbsReal::setPlotBins())

 The optional cut string expression can be used to select the events to be plotted.
 The cut specification may refer to any variable contained in the data set

 The drawOptions are passed to the TH1::Draw() method

 Create and fill a histogram with the asymmetry N[+] - N[-] / ( N[+] + N[-] ),
 where N(+/-) is the number of data points with asymCat=+1 and asymCat=-1
 as function of the frames variable. The asymmetry category 'asymCat' must
 have exactly 2 (or 3) states defined with index values +1,-1 (and 0)

 The plot range and the number of plot bins is determined by the parameters
 of the plot variable of the frame (RooAbsReal::setPlotRange(), RooAbsReal::setPlotBins())

 The optional cut string expression can be used to select the events to be plotted.
 The cut specification may refer to any variable contained in the data set

 The drawOptions are passed to the TH1::Draw() method

 Create and fill a histogram with the effiency N[1] / ( N[1] + N[0] ),
 where N(1/0) is the number of data points with effCat=1 and effCat=0
 as function of the frames variable. The efficiency category 'effCat' must
 have exactly 2 +1 and 0.

 The plot range and the number of plot bins is determined by the parameters
 of the plot variable of the frame (RooAbsReal::setPlotRange(), RooAbsReal::setPlotBins())

 The optional cut string expression can be used to select the events to be plotted.
 The cut specification may refer to any variable contained in the data set

 The drawOptions are passed to the TH1::Draw() method

 Loop over columns of our tree data and fill the input histogram. Returns a pointer to the
 input histogram, or zero in case of an error. The input histogram can be any TH1 subclass, and
 therefore of arbitrary dimension. Variables are matched with the (x,y,...) dimensions of the input
 histogram according to the order in which they appear in the input plotVars list.

 Create and fill a 1-dimensional table for given category column
 This functions is the equivalent of plotOn() for category dimensions.

 The optional cut string expression can be used to select the events to be tabulated
 The cut specification may refer to any variable contained in the data set

 The option string is currently not used

 Calculate standardized moment < (X - <X>)^n > / sigma^n,  where n = order.

 If cutSpec and/or cutRange are specified
 the moment is calculated on the subset of the data which pass the C++ cut specification expression 'cutSpec'
 and/or are inside the range named 'cutRange'

 Calculate moment < (X - <X>)^n > where n = order.

 If cutSpec and/or cutRange are specified
 the moment is calculated on the subset of the data which pass the C++ cut specification expression 'cutSpec'
 and/or are inside the range named 'cutRange'

 Return the 'order'-ed moment of observable 'var' in this dataset. If offset is non-zero it is subtracted
 from the values of 'var' prior to the moment calculation. If cutSpec and/or cutRange are specified
 the moment is calculated on the subset of the data which pass the C++ cut specification expression 'cutSpec'
 and/or are inside the range named 'cutRange'

 Create a RooRealVar containing the mean of observable 'var' in
 this dataset.  If cutSpec and/or cutRange are specified the
 moment is calculated on the subset of the data which pass the C++
 cut specification expression 'cutSpec' and/or are inside the
 range named 'cutRange'

 Create a RooRealVar containing the RMS of observable 'var' in
 this dataset.  If cutSpec and/or cutRange are specified the
 moment is calculated on the subset of the data which pass the C++
 cut specification expression 'cutSpec' and/or are inside the
 range named 'cutRange'

 Fill Doubles 'lowest' and 'highest' with the lowest and highest value of
 observable 'var' in this dataset. If the return value is kTRUE and error
 occurred

 Add a box with statistics information to the specified frame. By default a box with the
 event count, mean and rms of the plotted variable is added.

 The following optional named arguments are accepted

   What(const char* whatstr)          -- Controls what is printed: "N" = count, "M" is mean, "R" is RMS.
   Format(const char* optStr)         -- Classing [arameter formatting options, provided for backward compatibility
   Format(const char* what,...)       -- Parameter formatting options, details given below
   Label(const chat* label)           -- Add header label to parameter box
   Layout(Double_t xmin,              -- Specify relative position of left,right side of box and top of box. Position of
       Double_t xmax, Double_t ymax)     bottom of box is calculated automatically from number lines in box
   Cut(const char* expression)        -- Apply given cut expression to data when calculating statistics
   CutRange(const char* rangeName)    -- Only consider events within given range when calculating statistics. Multiple
                                         CutRange() argument may be specified to combine ranges

 The Format(const char* what,...) has the following structure

   const char* what          -- Controls what is shown. "N" adds name, "E" adds error,
                                "A" shows asymmetric error, "U" shows unit, "H" hides the value
   FixedPrecision(int n)     -- Controls precision, set fixed number of digits
   AutoPrecision(int n)      -- Controls precision. Number of shown digits is calculated from error
                                + n specified additional digits (1 is sensible default)
   VerbatimName(Bool_t flag) -- Put variable name in a \verb+   + clause.

 Implementation back-end of statOn() mehtod with named arguments

 Return the number of entries in this dataset

 Detailed printing interface

 Prepare dataset for use with cached constant terms listed in
 'cacheList' of expression 'arg'. Deactivate tree branches
 for any dataset observable that is either not used at all,
 or is used exclusively by cached branch nodes.

 Utility function that determines if all clients of object 'var'
 appear in given list of cached nodes.

 Return true if currently loaded coordinate is considered valid within
 the current range definitions of all observables

 Load a given row of data

 TTree Branch buffer size contro