Analyzing Monte-Carlo data

There is a range of functions that allow the calculation of statistical properties of Monte-Carlo data.

In general, when running a simulation, the results are recorded with an Observable-class and stored in a HDF5 file. These files can then later be read and statistical properties can be calculated. These results can also be written back into the file. There are several examples on how to do this in the examples/alea folder.

Here is an overview of the functions available

where AverageType is typename average_type<ValueType>::type, Timeseries is one of mcdata<ValueType>, mctimeseries<ValueType> or mctimeseries_view<ValueType> and Scalar MCTimeseries is one of mctimeseries<double> or mctimeseries_view<double>

The objects mctimeseries<ValueType> and mctimeseries_view<ValueType> are essentially wrapped boost::shared_ptr's to the timeseries. while the constructor of the mctimeseries class copies the whole data, the constructor of the mctimeseries_view class only creates a reference. One can easily create views of timeseries by using the functions cut_head and cut_tail:

Mean

Mean