| Function |
Description |
| Wavg |
Wavg(a,w)
or
Wavg(m)
Returns the weighted average value for each columns of a matrix.
|
| Wstd |
Wstd(a,w)
or
Wstd(m)
Returns the weighted standard deviation for each columns of a
matrix. |
| WSx |
WSx(a,w)
or
WSx(m)
Returns the weighted sum of the columns of a matrix. |
| WSxx |
WSxx(a,w)
or
WSxx(m)
Returns the weighted sum of the squares of each value in the
columns of a matrix. |
| WSxy |
WSxy(a,w)
or
WSxy(m)
Returns the sum found by multiplying together the elements in
a row and their weights and then summing the result for each row. |
| Wnum |
Wnum(a,w)
or
Wnum(m)
Sum of the weights for each column.
The a matrix isn't really necessary but is maintained for
syntax congruence |
| Wmm |
Wmm(a,w)
or
Wmm(m)
Finds the Minimum, Maximum values for each column in a matrix.
|
| Wlsq |
Wlsq(a,w)
or
Wlsq(m)
Returns the weighted least squares linear fit of a set of points.
a is a 2 columns matrix of x,y values
w is a 1 column matrix of weights
m is a 3 columns matrix with both values and weights.
Wlsq returns a the 1x3 matrix {a,b,r} where
y = ax + b
is the regression line and r the correlation coefficient.
|
| Wexpand |
Wexpand(a,w)
or
Wexpand(m)
Returns the expanded version of a weighted matrix.
The expanded matrix is a matrix in which each row of the original
matrix is repeated as many times as the corresponding weight. This
has a sense only if the weights are the same for all the columns
and are integers. |
Usage Note for all Wstat
functions:
All these functions can be called in two different ways:
Wstat(a, w) or Wstat(m)
where:
a is the matrix of values
w is the matrix of weights
m is a matrix containing both values and weights
All the operations are carried on column wise.
For the Wstat(a, w) syntax the following applies:
a and w must have the same numbers of rows.
If w has the same number of columns of a then each
column of w is applied as weights to the corresponding column
of a so that the weighted data are a[i,j]*w[i,j].
If w doesn't have the same number of columns of a
then the first column of w is applied as weights to all the
columns of a so that the weighted data are a[i,j]*w[1,j].
The other columns of w are ignored.
For the Wstat(m) syntax the following applies:
If m has two or more columns than its last column is used
as weights for all the previous columns of m so that the
weighted data are m[i,j]*m[i,n] for j = 1 to n-1.
If m has only one column then all the weights are set to
1. |
| CorrCoeff |
CorrCoeff(y1,y2,w)
or
CorrCoeff({y1 | y2 | w})
Returns the correlation coefficient for two optionally weighted
datasets.
A second coefficient is always computed ignoring the weights.
y1 = 1st set of y values
y2 = 2nd set of y values
w = optional weights |
| NormProbPlot |
NormProbPlot(y1,y2,w)
Computes the Least Squares Linear Fit for the Normal Probability
Plot of the given points.
If plotted plots the plot itself and the regression line.
A high correlation coefficient gives a reasonable confidence on
the normality of the points distribution.
If only one set of y is given the plot is computed on them. If y1
and y2 are given, the plot is computed on y2-y1 (residuals). If
also w is given the plot is computed on (y2-y1)*w |
| UnifProbPlot |
UnifProbPlot(y1,y2,w)
Computes the Least Squares Linear Fit for the Uniform Probability
Plot of the given points.
If plotted plots the plot itself and the regression line.
A high correlation coefficient gives a reasonable confidence on
the uniformity of the points distribution.
If only one set of y is given the plot is computed on them. If y1
and y2 are given, the plot is computed on y2-y1 (residuals). If
also w is given the plot is computed on (y2-y1)*w |
| RND |
RND(n,seed)
Returns a n rows vector of pseudo-random numbers uniformly distributed
in the interval [0-1].
Seed can be any positive integer. With the same seed the same sequence
is obtained. If seed is omitted a default sequence is produced.
Reference:
UNIRAN.f, part of the DATAPAC collection, which is © of NIST and
available at
http://www.itl.nist.gov/ div898/ software/datapac/homepage.htm |
| RNDNORM |
RNDNORM(n,seed)
Returns a n rows vector of pseudo-random numbers normally distributed
with standard deviation = 1 and mean = 0.
Seed can be any positive integer. With the same seed the same sequence
is obtained. If seed is omitted a default sequence is produced.
Reference:
NORRAN.f, part of the DATAPAC collection, which is © of NIST and
available at
http://www.itl.nist.gov/ div898/ software/datapac/homepage.htm |
| QuickSort |
QuickSort(dataInp, sortKey, sortMode)
Performs a quicksort on the input data.
dataInp = input data (vector or matrix)
sortKey = row or column to sort (optional, default 1)
sortMod = sort mode (optional, default 1):
1 : columnwise, ascending
2 : columnwise , descending
3 : rowwise, left to right
4 : rowwise, right to left |