Breaking update. It seriously changes parts of API. It adds new data types for
dealing with with estimates, confidence intervals, confidence levels and
p-value. Also API for statistical tests is changed.
-
Module Statistis.Types
now contains new data types for estimates,
upper/lower bounds, confidence level, and p-value.
CL
for representing confidence level
PValue
for representing p-values
Estimate
data type moved here from Statistis.Resampling.Bootstrap
and
now parametrized by type of error.
NormalError
— represents normal error.
ConfInt
— generic confidence interval
UpperLimit
,LowerLimit
for upper/lower limits.
-
New API for statistical tests. Instead of simply return significant/not
significant it returns p-value, test statistics and distribution of test
statistics if it’s available. Tests also return Nothing
instead of throwing
error if sample size is not sufficient. Fixes #25.
-
Statistics.Tests.Types.TestType
data type dropped
-
New smart constructors for distributions are added. They return Nothing
if
parameters are outside of allowed range.
-
Serialization instances (Show/Read, Binary, ToJSON/FromJSON
) for
distributions no longer allows to create data types with invalid
parameters. They will fail to parse. Cached values are not serialized either
so Binary
instances changed normal and F-distributions.
Encoding to JSON changed for Normal, F-distribution, and χ²
distributions. However data created using older statistics will be
successfully decoded.
Fixes #59.
-
Statistics.Resample.Bootstrap uses new data types for central estimates.
-
Function for calculation of confidence intervals for Poisson and binomial
distribution added in Statistics.ConfidenceInt
-
Tests of position now allow to ask whether first sample on average larger
than second, second larger than first or whether they differ significantly.
Affects Wilcoxon-T, Mann-Whitney-U, and Student-T tests.
-
API for bootstrap changed. New data types added.
-
Bug fixes for #74, #81, #83, #92, #94
-
complCumulative
added for many distributions.
-
The type classes Mean and Variance are split in two. This is
required for distributions which do not have finite variance or
mean.
-
The S.Sample.KernelDensity module has been renamed, and
completely rewritten to be much more robust. The older module
oversmoothed multi-modal data. (The older module is still
available under the name S.Sample.KernelDensity.Simple).
-
Histogram computation is added, in S.Sample.Histogram.
-
Discrete Fourie transform is added, in S.Transform
-
Root finding is added, in S.Math.RootFinding.
-
The complCumulative function is added to the Distribution
class in order to accurately assess probabilities P(X>x) which are
used in one-tailed tests.
-
A stdDev function is added to the Variance class for
distributions.
-
The constructor S.Distribution.normalDistr now takes standard
deviation instead of variance as its parameter.
-
A bug in S.Quantile.weightedAvg is fixed. It produced a wrong
answer if a sample contained only one element.
-
Bugs in quantile estimations for chi-square and gamma distribution
are fixed.
-
Integer overflow in mannWhitneyUCriticalValue is fixed. It
produced incorrect critical values for moderately large
samples. Something around 20 for 32-bit machines and 40 for 64-bit
ones.
-
A bug in mannWhitneyUSignificant is fixed. If either sample was
larger than 20, it produced a completely incorrect answer.
-
One- and two-tailed tests in S.Tests.NonParametric are selected
with sum types instead of Bool.
-
Test results returned as enumeration instead of Bool
.
-
Performance improvements for Mann-Whitney U and Wilcoxon tests.
-
Module S.Tests.NonParamtric
is split into S.Tests.MannWhitneyU
and S.Tests.WilcoxonT
-
sortBy is added to S.Function.
-
Mean and variance for gamma distribution are fixed.
-
Much faster cumulative probability functions for Poisson and
hypergeometric distributions.
-
Better density functions for gamma and Poisson distributions.
-
Student-T, Fisher-Snedecor F-distributions and Cauchy-Lorentz
distribution are added.
-
The function S.Function.create is removed. Use generateM from
the vector package instead.
-
Function to perform approximate comparison of doubles is added to
S.Function.Comparison
-
Regularized incomplete beta function and its inverse are added to
S.Function