Procedures to do statistics. More...

Functions/Subroutines
real(double) function	median (data, mask)
	Compute the median of a data array.

real function	median_sp (data, mask)
	Compute the median of a data array - single-precision wrapper for median()

real(double) function	mean (data, mask)
	Compute the mean of a data array.

real function	mean_sp (data, mask)
	Compute the mean of a data array - single-precision wrapper for mean()

real(double) function	mean_weight (data, wgts, mask)
	Compute the weighted mean of a data array.

real(double) function	stdev (data, dmean, mask, var)
	Compute the standard deviation of a data array with mean 'mean'.

real function	stdev_sp (data, mean, mask)
	Compute the standard deviation of a data array with mean 'mean' - single-precision wrapper for stDev()

real(double) function	stdev_weight (data, wgts, wmean, mask)
	Compute the weighted standard deviation of a data array with weighted mean 'wMean'.

pure subroutine	mean_var_running (mean, var, data, num, stdev)
	Compute a running mean and variance by adding a data point and the data point number to existing values. If num=1, initialise. Note that mean and var are I/O variables and cannot be dummy variables or values. Num must be accurate, and increased by one between calls by the user. Optionally, return the standard deviation.

subroutine	prob_range (data, range, llim, ulim, mask)
	Find a given probability range for a data array - the narrowest range that contains a given fraction of data points.

subroutine	prob_range_sp (data, range, llim, ulim, mask)
	Find a given probability range for a data array - the narrowest range that contains a given fraction of data points - single-precision wrapper for prob_range()

pure integer function	determine_nbin_1d (npoints)
	Roughly estimate the number of 1D bins needed, from the number of data points.

subroutine	bin_data_1d (xdat, nbin, norm, mode, cumul, xmin, xmax, xbin, ybin)
	Bin data in 1D bins by counting the number of data points in each bin.

subroutine	bin_data_1d_sp (xdat, nbin, norm, mode, cumul, xmin, xmax, xbin, ybin)
	Bin data in 1D bins by counting the number of data points in each bin - single-precision wrapper for bin_data_1d()

pure subroutine	histogram_1d_onthefly (xdat, nbin, mode, cumul, xmin, xmax, xbin, ybin, init, weight)
	Create a 1D histogram on the fly (point by point). Bin data points by computing the bin they should be in.

subroutine	bin_data_2d (xdat, ydat, norm, nxbin, nybin, xmin, xmax, ymin, ymax, z, tr, weights)
	Bin data in 2 dimensions - computing the bin number rather than searching for it is ~10x faster.

subroutine	bin_data_2d_sp (xdat, ydat, norm, nxbin, nybin, xmin, xmax, ymin, ymax, z, tr, weights)
	Bin data in 2 dimensions - single-precision wrapper for bin_data_2d()

subroutine	histogram_2d_onthefly (xdat, ydat, nxbin, nybin, xmin, xmax, ymin, ymax, z, init, weight, tr)
	Bin data in 2 dimensions - computing the bin number rather than searching for it is ~10x faster.

pure integer(long) function	faculty_i (n)
	Compute the faculty of an integer, returning a long integer.

pure real(double) function	faculty (n)
	Compute the faculty of an integer, returning a double-precision real.

pure real(double) function	binom_coeff (n, k)
	Compute the binomial coefficient of n and k.

pure real(double) function	binom_prob (n, k, p)
	Compute the binomial probability of n and k, and probability p.

pure real(double) function	binom_cumul_prob (n, k, p)
	Compute the cumulative binomial probability of n and k OR FEWER, and probability p.

pure real(double) function	poisson_prob (k, lambda)
	Compute the Poisson probability of EXACTLY k events occurring in a fixed interval for a known average rate lambda, and independently of the time since the last event: P = λ^k e^-λ / k!

pure real(double) function	poisson_prob_cumul (k, lambda)
	Compute the cumulative Poisson probability of k OR FEWER events occurring in a fixed interval for a known average rate lambda, and independently of the time since the last event: P = Σ(i=0,k) λ^i e^-λ / i!

real(double) function	correlation (data1, data2)
	Compute the normalised correlation between two data series.

Detailed Description

Procedures to do statistics.

Function/Subroutine Documentation

◆ bin_data_1d()

subroutine sufr_statistics::bin_data_1d	(	real(double), dimension(:), intent(in)	xdat,
		integer, intent(in)	nbin,
		logical, intent(in)	norm,
		integer, intent(in)	mode,
		logical, intent(in)	cumul,
		real(double), intent(inout)	xmin,
		real(double), intent(inout)	xmax,
		real(double), dimension(:), intent(inout)	xbin,
		real(double), dimension(:), intent(inout)	ybin )

Bin data in 1D bins by counting the number of data points in each bin.

Parameters

xDat	Data to be binned (ndat points)
Nbin	Desired number of bins. Note that the binned-data arrays xbin and ybin must have size >= Nbin+1
norm	Normalise histogram (T) or not (F)
mode	Mode: -1: xbin is left of bin, 0: xbin is centre of bin, 1: xbin is right of bin
cumul	Make a cumulative histogram (T/F)
xMin	Minimum value of the binning range. Set xMin=xMax to auto-determine (I/O)
xMax	Maximum value of the binning range. Set xMin=xMax to auto-determine (I/O)
xBin	Binned data, location of the bins. The x values are the left side of the bin! (output)
yBin	Binned data, height of the bins. I/O so that the array size can be checked (output)

Definition at line 596 of file statistics.f90.

References sufr_kinds::double, and sufr_system::quit_program_error().

Referenced by bin_data_1d_sp().

Here is the call graph for this function:

◆ bin_data_1d_sp()

subroutine sufr_statistics::bin_data_1d_sp	(	real, dimension(:), intent(in)	xdat,
		integer, intent(in)	nbin,
		logical, intent(in)	norm,
		integer, intent(in)	mode,
		logical, intent(in)	cumul,
		real, intent(inout)	xmin,
		real, intent(inout)	xmax,
		real, dimension(:), intent(inout)	xbin,
		real, dimension(:), intent(inout)	ybin )

Bin data in 1D bins by counting the number of data points in each bin - single-precision wrapper for bin_data_1d()

Parameters

xDat	Data to be binned (ndat points)
Nbin	Desired number of bins. Note that the binned-data arrays xBin and yBin must have size >= Nbin+1
norm	Normalise histogram (T) or not (F)
mode	Mode: -1: xBin is left of bin, 0: xBin is centre of bin, 1: xBin is right of bin
cumul	Make a cumulative histogram (T/F)
xMin	Minimum value of the binning range. Set xMin=xMax to auto-determine (I/O)
xMax	Maximum value of the binning range. Set xMin=xMax to auto-determine (I/O)
xBin	Binned data, location of the bins. The x values are the left side of the bin! (output)
yBin	Binned data, height of the bins. I/O so that the array size can be checked (output)

Definition at line 670 of file statistics.f90.

References bin_data_1d(), sufr_kinds::double, and sufr_system::quit_program_error().

Here is the call graph for this function:

◆ bin_data_2d()

subroutine sufr_statistics::bin_data_2d	(	real(double), dimension(:), intent(in)	xdat,
		real(double), dimension(:), intent(in)	ydat,
		integer, intent(in)	norm,
		integer, intent(in)	nxbin,
		integer, intent(in)	nybin,
		real(double), intent(inout)	xmin,
		real(double), intent(inout)	xmax,
		real(double), intent(inout)	ymin,
		real(double), intent(inout)	ymax,
		real(double), dimension(nxbin+1,nybin+1), intent(out)	z,
		real(double), dimension(6), intent(out)	tr,
		real(double), dimension(size(xdat)), intent(in), optional	weights )

Bin data in 2 dimensions - computing the bin number rather than searching for it is ~10x faster.

Parameters

xDat	Input data: x values - array size: ndat
yDat	Input data: y values - array size: ndat
norm	Normalise the bins (1) or not (0)
nxBin	Desired number of bins in the x direction
nyBin	Desired number of bins in the y direction
xMin	Lower limit for the binning range in the x direction - autodetermine if xMin = xMax
xMax	Upper limit for the binning range in the x direction - autodetermine if xMin = xMax
yMin	Lower limit for the binning range in the y direction - autodetermine if yMin = yMax
yMax	Upper limit for the binning range in the y direction - autodetermine if yMin = yMax
z	Binned data set z(nxBin+1,nyBin+1) - this array may be larger than you expect - nbin bins have nbin+1 borders (output)
tr	Transformation elements for pgplot tr(6) (output)
weights	Weights to use when binning data, same size as xDat,yDat (optional)

Definition at line 804 of file statistics.f90.

References sufr_kinds::double, and sufr_system::quit_program_error().

Referenced by bin_data_2d_sp().

Here is the call graph for this function:

◆ bin_data_2d_sp()

subroutine sufr_statistics::bin_data_2d_sp	(	real, dimension(:), intent(in)	xdat,
		real, dimension(:), intent(in)	ydat,
		integer, intent(in)	norm,
		integer, intent(in)	nxbin,
		integer, intent(in)	nybin,
		real, intent(inout)	xmin,
		real, intent(inout)	xmax,
		real, intent(inout)	ymin,
		real, intent(inout)	ymax,
		real, dimension(nxbin+1,nybin+1), intent(out)	z,
		real, dimension(6), intent(out)	tr,
		real, dimension(size(xdat)), intent(in), optional	weights )

Bin data in 2 dimensions - single-precision wrapper for bin_data_2d()

Parameters

xDat	Input data: x values - array size: ndat
yDat	Input data: y values - array size: ndat
norm	Normalise the bins (1) or not (0)
nxBin	Desired number of bins in the x direction
nyBin	Desired number of bins in the y direction
xMin	Lower limit for the binning range in the x direction - autodetermine if xMin = xMax
xMax	Upper limit for the binning range in the x direction - autodetermine if xMin = xMax
yMin	Lower limit for the binning range in the y direction - autodetermine if yMin = yMax
yMax	Upper limit for the binning range in the y direction - autodetermine if yMin = yMax
z	Binned data set z(nxBin+1,nyBin+1) - this array may be larger than you expect - nbin bins have nbin+1 borders (output)
tr	Transformation elements for pgplot tr(6) (output)
weights	Weights to use when binning data, same size as xDat,yDat (optional)

Definition at line 906 of file statistics.f90.

References bin_data_2d(), sufr_kinds::double, and sufr_system::quit_program_error().

Here is the call graph for this function:

◆ binom_coeff()

pure real(double) function sufr_statistics::binom_coeff	(	integer, intent(in)	n,
		integer, intent(in)	k )

Compute the binomial coefficient of n and k.

Parameters

n	Total number of trials; n in "n choose k"
k	Number of succesful trials; k in "n choose k"

Return values

binom_coeff Binomial coefficient n! / [k!(n-k)!]

See also: https://en.wikipedia.org/wiki/Binomial_coefficient#Binomial_coefficient_in_programming_languages

Definition at line 1080 of file statistics.f90.

References binom_coeff(), and sufr_kinds::double.

Referenced by binom_coeff(), and binom_prob().

Here is the call graph for this function:

◆ binom_cumul_prob()

pure real(double) function sufr_statistics::binom_cumul_prob	(	integer, intent(in)	n,
		integer, intent(in)	k,
		real(double), intent(in)	p )

Compute the cumulative binomial probability of n and k OR FEWER, and probability p.

Parameters

n	Total number of trials; n in "n choose k"
k	Number of succesful trials OR FEWER; k in "n choose k"
p	Probability of a succesful trial

Return values

binom_cumul_prob Cumulative binomial probability Σ(i=0,k) n! / [i!(n-i)!] * p^i * (1-p)^(n-i)

Definition at line 1140 of file statistics.f90.

References binom_cumul_prob(), binom_prob(), and sufr_kinds::double.

Referenced by binom_cumul_prob().

Here is the call graph for this function:

◆ binom_prob()

pure real(double) function sufr_statistics::binom_prob	(	integer, intent(in)	n,
		integer, intent(in)	k,
		real(double), intent(in)	p )

Compute the binomial probability of n and k, and probability p.

Parameters

n	Total number of trials; n in "n choose k"
k	Number of succesful trials; k in "n choose k"
p	Probability of a succesful trial

Return values

binom_prob Binomial probability n! / [k!(n-k)!] * p^k * (1-p)^(n-k)

Definition at line 1117 of file statistics.f90.

References binom_coeff(), binom_prob(), and sufr_kinds::double.

Referenced by binom_cumul_prob(), and binom_prob().

Here is the call graph for this function:

◆ correlation()

real(double) function sufr_statistics::correlation	(	real(double), dimension(:), intent(in)	data1,
		real(double), dimension(:), intent(in)	data2 )

Compute the normalised correlation between two data series.

Parameters

data1	Data series 1
data2	Data series 2 - should have the same length as data1

Return values

correlation Normalised correlation [-1,1] between the two data series

Definition at line 1217 of file statistics.f90.

References correlation(), sufr_kinds::double, mean(), and sufr_system::warn().

Referenced by correlation().

Here is the call graph for this function:

◆ determine_nbin_1d()

pure integer function sufr_statistics::determine_nbin_1d ( integer, intent(in) npoints )

Roughly estimate the number of 1D bins needed, from the number of data points.

Parameters

npoints Number of data points

Return values

determine_nbin_1d Number of bins

Definition at line 564 of file statistics.f90.

References determine_nbin_1d().

Referenced by determine_nbin_1d().

Here is the call graph for this function:

◆ faculty()

pure real(double) function sufr_statistics::faculty ( integer, intent(in) n )

Compute the faculty of an integer, returning a double-precision real.

Parameters

n	Number - can be up to 170 for double-precision reals (as opposed to 20 for long integers and 13 for integers)

Return values

faculty Faculty of n; n! (double-precision real)

Definition at line 1055 of file statistics.f90.

References sufr_kinds::double, and faculty().

Referenced by faculty(), and poisson_prob().

Here is the call graph for this function:

◆ faculty_i()

pure integer(long) function sufr_statistics::faculty_i ( integer, intent(in) n )

Compute the faculty of an integer, returning a long integer.

Parameters

n	Number - up to 20 for long integers (up to 13 for integers)

Return values

faculty_i Faculty of n; n! - a long integer

Definition at line 1033 of file statistics.f90.

References faculty_i(), and sufr_kinds::long.

Referenced by faculty_i().

Here is the call graph for this function:

◆ histogram_1d_onthefly()

pure subroutine sufr_statistics::histogram_1d_onthefly	(	real(double), intent(in)	xdat,
		integer, intent(in)	nbin,
		integer, intent(in)	mode,
		logical, intent(in)	cumul,
		real(double), intent(in)	xmin,
		real(double), intent(in)	xmax,
		real(double), dimension(:), intent(inout)	xbin,
		real(double), dimension(:), intent(inout)	ybin,
		logical, intent(in), optional	init,
		real(double), intent(in), optional	weight )

Create a 1D histogram on the fly (point by point). Bin data points by computing the bin they should be in.

Note: Call this routine once to initialise (init=1; set xBin), then call many times to collect data and construct yBin (init=0)

Parameters

xDat	Data point to be binned
Nbin	Desired number of bins. Note that the binned-data arrays xBin and yBin must have size >= Nbin+1
mode	Mode: -1: xBin is left of bin, 0: xBin is centre of bin, 1: xBin is right of bin
cumul	Make a cumulative histogram (T/F)
xMin	Minimum value of the binning range. Set xMin=xMax to auto-determine (I/O)
xMax	Maximum value of the binning range. Set xMin=xMax to auto-determine (I/O)
xBin	Binned data, location of the bins. The x values are the left side of the bin! (output)
yBin	Binned data, height of the bins. I/O so that the array size can be checked (output)
init	Initialisation call: true/false (data collection). Optional; default=false.
weight	Add weight to the bin, rather than 1. Optional; default=1.

Definition at line 719 of file statistics.f90.

References sufr_kinds::double, and sufr_system::quit_program_error().

Here is the call graph for this function:

◆ histogram_2d_onthefly()

subroutine sufr_statistics::histogram_2d_onthefly	(	real(double), intent(in)	xdat,
		real(double), intent(in)	ydat,
		integer, intent(in)	nxbin,
		integer, intent(in)	nybin,
		real(double), intent(in)	xmin,
		real(double), intent(in)	xmax,
		real(double), intent(in)	ymin,
		real(double), intent(in)	ymax,
		real(double), dimension(nxbin+1,nybin+1), intent(out)	z,
		logical, intent(in), optional	init,
		real(double), intent(in), optional	weight,
		real(double), dimension(6), intent(out), optional	tr )

Bin data in 2 dimensions - computing the bin number rather than searching for it is ~10x faster.

Parameters

xDat	Input data point: x value
yDat	Input data point: y value
nxBin	Desired number of bins in the x direction
nyBin	Desired number of bins in the y direction
xMin	Lower limit for the binning range in the x direction - autodetermine if xMin = xMax
xMax	Upper limit for the binning range in the x direction - autodetermine if xMin = xMax
yMin	Lower limit for the binning range in the y direction - autodetermine if yMin = yMax
yMax	Upper limit for the binning range in the y direction - autodetermine if yMin = yMax
z	Binned data set z(nxBin+1,nyBin+1) - this array may be larger than you expect - nbin bins have nbin+1 borders (output)
init	Init mode: true/false (optional)
weight	Weight to use when binning data, same size as xDat,yDat (optional)
tr	Transformation elements for pgplot tr(6) (output; optional)

Definition at line 967 of file statistics.f90.

References sufr_kinds::double, and sufr_system::quit_program_error().

Here is the call graph for this function:

◆ mean()

real(double) function sufr_statistics::mean	(	real(double), dimension(:), intent(in)	data,
		logical, dimension(:), intent(in), optional	mask )

Compute the mean of a data array.

Parameters

data	1D array of data points
mask	Mask to apply to data (optional)

Return values

mean	The mean of a data array

Definition at line 119 of file statistics.f90.

References sufr_kinds::double, mean(), sufr_system::quit_program_error(), and sufr_system::warn().

Referenced by correlation(), mean(), mean_sp(), mean_var_running(), stdev(), and stdev_sp().

Here is the call graph for this function:

◆ mean_sp()

real function sufr_statistics::mean_sp	(	real, dimension(:), intent(in)	data,
		logical, dimension(:), intent(in), optional	mask )

Compute the mean of a data array - single-precision wrapper for mean()

Parameters

data	1D array of data points
mask	Mask to apply to data (optional)

Return values

mean_sp The mean of a data array

Definition at line 158 of file statistics.f90.

References sufr_kinds::double, mean(), mean_sp(), and sufr_system::quit_program_error().

Referenced by mean_sp().

Here is the call graph for this function:

◆ mean_var_running()

pure subroutine sufr_statistics::mean_var_running	(	real(double), intent(inout)	mean,
		real(double), intent(inout)	var,
		real(double), intent(in)	data,
		integer, intent(in)	num,
		real(double), intent(out), optional	stdev )

Compute a running mean and variance by adding a data point and the data point number to existing values. If num=1, initialise. Note that mean and var are I/O variables and cannot be dummy variables or values. Num must be accurate, and increased by one between calls by the user. Optionally, return the standard deviation.

Parameters

mean	Running mean (I/O)
var	Running variance (I/O)
data	New/current data point
num	Number of the current data point
stDev	Current standard deviation (output; optional)

See also: https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance#Na%C3%AFve_algorithm

Definition at line 408 of file statistics.f90.

References sufr_kinds::double, mean(), and stdev().

Here is the call graph for this function:

◆ mean_weight()

real(double) function sufr_statistics::mean_weight	(	real(double), dimension(:), intent(in)	data,
		real(double), dimension(:), intent(in)	wgts,
		logical, dimension(:), intent(in), optional	mask )

Compute the weighted mean of a data array.

Parameters

data	1D array of data points
wgts	Weights for all data points
mask	Mask to apply to data (optional)

Return values

mean_weight The weighted mean of a data array

See also: https://en.wikipedia.org/wiki/Weighted_arithmetic_mean

Definition at line 196 of file statistics.f90.

References sufr_kinds::double, sufr_system::error(), mean_weight(), and sufr_system::quit_program_error().

Referenced by mean_weight(), and stdev_weight().

Here is the call graph for this function:

◆ median()

real(double) function sufr_statistics::median	(	real(double), dimension(:), intent(in)	data,
		logical, dimension(:), intent(in), optional	mask )

Compute the median of a data array.

Parameters

data	1D array of data points
mask	Mask to apply to data (optional)

Return values

median The median of a data array

Definition at line 37 of file statistics.f90.

References sufr_kinds::double, median(), sufr_system::quit_program_error(), and sufr_sorting::sorted_index_list().

Referenced by median(), and median_sp().

Here is the call graph for this function:

◆ median_sp()

real function sufr_statistics::median_sp	(	real, dimension(:), intent(in)	data,
		logical, dimension(:), intent(in), optional	mask )

Compute the median of a data array - single-precision wrapper for median()

Parameters

data	1D array of data points
mask	Mask to apply to data (optional)

Return values

median_sp The median of a data array

Definition at line 85 of file statistics.f90.

References sufr_kinds::double, median(), median_sp(), and sufr_system::quit_program_error().

Referenced by median_sp().

Here is the call graph for this function:

◆ poisson_prob()

pure real(double) function sufr_statistics::poisson_prob	(	integer, intent(in)	k,
		real(double), intent(in)	lambda )

Compute the Poisson probability of EXACTLY k events occurring in a fixed interval for a known average rate lambda, and independently of the time since the last event: P = λ^k e^-λ / k!

Parameters

k	Number of events
lambda	Average event rate

Return values

poisson_prob Poisson probability P = λ^k e^-λ / k!

Definition at line 1167 of file statistics.f90.

References sufr_kinds::double, faculty(), and poisson_prob().

Referenced by poisson_prob(), and poisson_prob_cumul().

Here is the call graph for this function:

◆ poisson_prob_cumul()

pure real(double) function sufr_statistics::poisson_prob_cumul	(	integer, intent(in)	k,
		real(double), intent(in)	lambda )

Compute the cumulative Poisson probability of k OR FEWER events occurring in a fixed interval for a known average rate lambda, and independently of the time since the last event: P = Σ(i=0,k) λ^i e^-λ / i!

Parameters

k	Number of events
lambda	Average event rate

Return values

poisson_prob_cumul Cumulative Poisson probability P = Σ(i=0,k) λ^i e^-λ / i!

Definition at line 1191 of file statistics.f90.

References sufr_kinds::double, poisson_prob(), and poisson_prob_cumul().

Referenced by poisson_prob_cumul().

Here is the call graph for this function:

◆ prob_range()

subroutine sufr_statistics::prob_range	(	real(double), dimension(:), intent(in)	data,
		real(double), intent(in)	range,
		real(double), intent(out)	llim,
		real(double), intent(out)	ulim,
		logical, dimension(:), intent(in), optional	mask )

Find a given probability range for a data array - the narrowest range that contains a given fraction of data points.

Parameters

data	1D array of data points
range	Probability range - e.g. 0.95 = 95% probability ~ "2-sigma"
llim	Lower limit of probability range (output)
ulim	Upper limit of probability range (output)
mask	Mask to apply to data (optional)

Definition at line 447 of file statistics.f90.

References sufr_numerics::deq(), sufr_kinds::double, sufr_system::error(), sufr_system::quit_program_error(), sufr_sorting::sorted_index_list(), and sufr_system::warn().

Referenced by prob_range_sp().

Here is the call graph for this function:

◆ prob_range_sp()

subroutine sufr_statistics::prob_range_sp	(	real, dimension(:), intent(in)	data,
		real, intent(in)	range,
		real, intent(out)	llim,
		real, intent(out)	ulim,
		logical, dimension(:), intent(in), optional	mask )

Find a given probability range for a data array - the narrowest range that contains a given fraction of data points - single-precision wrapper for prob_range()

Parameters

data	1D array of data points
range	Probability range - e.g. 0.95 = 95% probability ~ "2-sigma"
llim	Lower limit of probability range (output)
ulim	Upper limit of probability range (output)
mask	Mask to apply to data (optional)

Definition at line 521 of file statistics.f90.

References sufr_kinds::double, prob_range(), sufr_system::quit_program_error(), and sufr_sorting::sorted_index_list().

Here is the call graph for this function:

◆ stdev()

real(double) function sufr_statistics::stdev	(	real(double), dimension(:), intent(in)	data,
		real(double), intent(in), optional	dmean,
		logical, dimension(:), intent(in), optional	mask,
		real(double), intent(out), optional	var )

Compute the standard deviation of a data array with mean 'mean'.

Parameters

data	1D array with data points
dMean	Mean of the data points (optional; will be computed if not provided)
mask	Mask to apply to data (optional)
var	Variance of the data (output)

Return values

stDev The standard deviation

See also: https://en.wikipedia.org/wiki/Standard_deviation

Definition at line 250 of file statistics.f90.

References sufr_kinds::double, sufr_system::error(), mean(), sufr_system::quit_program_error(), and stdev().

Referenced by mean_var_running(), stdev(), and stdev_sp().

Here is the call graph for this function:

◆ stdev_sp()

real function sufr_statistics::stdev_sp	(	real, dimension(:), intent(in)	data,
		real, intent(in)	mean,
		logical, dimension(:), intent(in), optional	mask )

Compute the standard deviation of a data array with mean 'mean' - single-precision wrapper for stDev()

Parameters

data	1D array with data points
mean	Mean of the data points
mask	Mask to apply to data (optional)

Return values

stDev The standard deviation

Definition at line 308 of file statistics.f90.

References sufr_kinds::double, mean(), sufr_system::quit_program_error(), stdev(), and stdev_sp().

Referenced by stdev_sp().

Here is the call graph for this function:

◆ stdev_weight()

real(double) function sufr_statistics::stdev_weight	(	real(double), dimension(:), intent(in)	data,
		real(double), dimension(:), intent(in)	wgts,
		real(double), intent(in), optional	wmean,
		logical, dimension(:), intent(in), optional	mask )

Compute the weighted standard deviation of a data array with weighted mean 'wMean'.

Parameters

data	1D array with data points
wgts	Weights for all data points
wMean	Weighted mean of the data points (optional)
mask	Mask to apply to data (optional)

Return values

stDev_weight The weighted standard deviation

Definition at line 343 of file statistics.f90.

References sufr_kinds::double, sufr_system::error(), mean_weight(), sufr_system::quit_program_error(), and stdev_weight().

Referenced by stdev_weight().

Here is the call graph for this function:

Functions/Subroutines

Detailed Description

Function/Subroutine Documentation

◆ bin_data_1d()

◆ bin_data_1d_sp()

◆ bin_data_2d()

◆ bin_data_2d_sp()

◆ binom_coeff()

◆ binom_cumul_prob()

◆ binom_prob()

◆ correlation()

◆ determine_nbin_1d()

◆ faculty()

◆ faculty_i()

◆ histogram_1d_onthefly()

◆ histogram_2d_onthefly()

◆ mean()

◆ mean_sp()

◆ mean_var_running()

◆ mean_weight()

◆ median()

◆ median_sp()

◆ poisson_prob()

◆ poisson_prob_cumul()

◆ prob_range()

◆ prob_range_sp()

◆ stdev()

◆ stdev_sp()

◆ stdev_weight()