Monte Carlo#
Monte Carlo simulations are a mathematical technique used to predict possible outcomes on an uncertain event. Pvdeg uses pv modeling constants as variables to preform Monte Carlo simulations. These parameters include but are not limited to R_0 Frequency factor, prefactor,
Add non-exhaustive list of possible monte carlo variables
See the PVdeg Monte Carlo tutorials for specific case details here (insert later)
General Steps#
Defining Correlations#
If your variables are correlated form a list of correlations using pvdeg.montecarlo.Corr() objects.
my_correlations = [pvdeg.montecarlo.corr('<var_1>', '<var_2>', <float>), ...]
Defining Mean and Standard Deviation#
Each variable passed to a Corr object in string form must have its own dictionary entry containing its mean and standard deviation in the following form.
`
my_dict = {
<var_1> : {'mean' : <float>, 'stdev' : <float>}
}
`
why are there extra backticks this is upsetting
Generating Correlated Samples#
pvdeg.montecarlo.generateCorrelatedSamples() can be used to create correlated samples for our monte carlo simulation. The image below shows how to provide its arguments in detail. Note: seed is an optional argument.
3rd Party Samples/Data#
If you have another prefered way of creating samples to use in a Monte Carlo simulation you can bypass the previous steps. The samples DataFrame must have appropriate column names that match the target function’s arguments.
Running A Monte Carlo Simulation#
Use pvdeg.montecarlo.simulate() to run a Monte Carlo simulation on a target function. See the image below for more detail.
Selecting Static and Dynamic Variables#
See the image below which shows how to split your Monte Carlo inputs into the appropriate function arguments.
