Mathworks STATISTICS TOOLBOX RELEASE NOTES

Statistics Toolbox™ Release Notes

How to Contact The MathWorks

www.mathworks. comp.soft-sys.matlab Newsgroup www.mathworks.com/contact_TS.html Technical Support

suggest@mathworks.com Product enhancement suggestions

bugs@mathwo doc@mathworks.com Documentation error reports service@mathworks.com Order status, license renewals, passcodes

info@mathwo

com

rks.com

Web

Bug reports

Sales, prici

ng, and general information

508-647-7000 (Phone)

508-647-7001 (Fax)

The MathWorks, Inc. 3 Apple Hill Drive Natick, MA 01760-2098

For contact information about worldwide offices, see the MathWorks Web site.

Statistics Toolbox™ Release Notes

The software described in this document is furnished under a license agreement. The software may be used or copied only under the terms of the license agreement. No part of this manual may be photocopied or reproduced in any form without prior written consent from The MathW orks, Inc.

FEDERAL ACQUISITION: This provision applies to all acquisitions of the Program and Documentation by, for, or through the federal government of the United States. By accepting delivery of the Program or Documentation, the government hereby agrees that this software or documentation qualifies as commercial computer software or commercial computer software documentation as such terms are used or defined in FAR 12.212, DFARS Part 227.72, and DFARS 252.227-7014. Accordingly, the terms and conditions of this Agreement and only those rights specified in this Agreement, shall pertain to and govern theuse,modification,reproduction,release,performance,display,anddisclosureoftheProgramand Documentation by the federal government (or other entity acquiring for or through the federal government) and shall supersede any conflicting contractual terms or conditions. If this License fails to meet the government’s needs or is inconsistent in any respect with federal procurement law, the government agrees to return the Program and Docu mentation, unused, to The MathWorks, Inc.

Trademarks

MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See

www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand

names may be trademarks or registered trademarks of their respective holders.

Patents

The MathWorks products are protected by one or more U.S. patents. Please see

www.mathworks.com/patents for more information.

Summary by Version ............................... 1

Contents

Version 7.3 (R2010a) Statistics Toolbox Software

Version 7.2 (R2009b) Statistics Toolbox Software

Version 7.1 (R2009a) Statistics Toolbox Software

Version 7.0 (R2008b) Statistics Toolbox Software

Version 6.2 (R2008a) Statistics Toolbox Software

Version 6.1 (R2007b) Statistics Toolbox Software

Version 6.0 (R2007a) Statistics Toolbox Software

Version 5.3 (R2006b) Statistics Toolbox Software

Version 5.2 (R2006a) Statistics Toolbox Software

Version 5.1 (R14SP3) Statistics Toolbox Software

..... 4

..... 6

..... 8

..... 11

..... 15

..... 17

..... 21

..... 25

..... 29

..... 33

Version 5.0.2 (R14SP2) Statistics Toolbox Software

Compatibility Summary for Statistics Toolbox

Software

........................................ 38

... 37

iii

iv Contents

SummarybyVersion

This table provides quick access to what’s new in each version. For clarification, see “Using Release Notes” on page 2 .

Statistics Toolbox™ Release Notes

Version (Release)

Latest Versi V7.3 (R2010a

V7.2 (R2009b)

V7.1 (R2009a)

V7.0 (R2

V6.2 (R2008a)

V6.1 (R2007b)

V6.0 (

V5.3 (R2006b)

008b)

R2007a)

New Features and Changes

Yes

)

Details

Yes Details

Yes Detai

Yes Details

Version Compatibilit Consideratio

No Bug Reports

No Bug Repor

Yes Summary

Yes

Summa

Yes Summary

Fixed Bugs

and Known

Problems

Includes fix

Includes fixes

Includes

No No

Bug Reports Includes fixes

Bug Rep Inclu

Bug Reports Includes fixes

orts

des fixes

fixes

Related Documentation at Web Site

Printable R elease Notes: PDF

Current product documentation

V5.2 (R2006a)

.1 (R14SP3)

V5.0.2 (R14SP2)

Yes

ails

Det

Yes Details

No Bug

No No No

No Bug Reports

Reports

ludes fixes

Inc

Includes fixes

Statistics Toolbox™ Release Notes

Using Release No

Use release note

• New features

• Changes

• Potential imp

Review the re product (for bugs, or comp

If you are up review the c you upgrad

What Is in t

New Featu

• New func

• Changes

s when upgrading to a new er version to learn about:

act o n your existing files and practices

lease notes for other M athWorks™ products required for this

example, MATLAB

atibility considerations in other products impact you.

grading from a softw are version other than the most recent one,

urrent release notes and all interim versions. For example, when

e from V1.0 to V1.2, review the release notes for V1.1 and V1.2.

he Release Notes

res and Changes

tionality

to existing functionality

tes

or Simulink®). Determine if enhancements,

Versio

When a n versi impac

Comp Repo in in comp

Fix

The vi

n Compatibility Considerations

ew feature or change introduces a reported incompatibil ity between

ons, the Compatibility Considerations subsection explains the

atibility issues reported after the product release appear under Bug rts at The MathWorks™ Web site. Bug fixes can sometimes result compatibilities, so review the fixed bugs in Bug Reports for any

atibility impact.

ed Bugs and Known Problems

MathWorks offers a user-searchable Bug Reports database so you can

ew Bug Reports. The development team updates this database at release

SummarybyVersion

time and as more information becomes available. Bug Reports include provisions for any known workarounds or file replacem ents. Information is available for bugs existing in or fixed in Release 14SP2 or later. Information is not avail able for all bugs in earlier releases.

Access Bug Reports using y our MathWorks Account.

Statistics Toolbox™ Release Notes

Version 7.3 (R2010a) Statistics Toolbox Software

This table summarizes what’s new in Version 7.3 (R2010a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports

New features and changes introduced in this version are:

• “Stochastic Algorithm Functionality in NLME Models” on page 4

• “k-Nearest Neighbor Searching” on page 4

• “Confidence Intervals Option in perfcurve” on page 4

• “Observation Weights Options in Resampling Functions” on page 5

Fixed Bugs an d Known Problems

Includes fixes

Related Documentation at Web Site

Printable Release Notes: PDF

Current product documentation

Stochastic Algorithm F unctionality in NLME Models

New stochastic algorithm for fitting NLME models is more robust with respect to starting values, enables parameter transformations, and relaxes assumption of constant error variance. See

nlmefitsa.

k-Nearest Neighbor Searching

New functions for k-Nearest Neighbor (kNN) search efficiently to find the closest points to any query point. For information, see “k-Nearest Neighbor Search”.

Confidence Intervals Option in perfcurve

Newoptionintheperfcu rve function to compute co n fide nce intervals for classifier performance curve s.

Version 7.3 (R2010a) Statistics Toolbox™ Software

Observation Wei Functions

New options to we supported by

ight resampling probabilities broadens the range of models

otstrp

, bootci,andperfcurve functions.

ghts Options in Resampling

Statistics Toolbox™ Release Notes

Version 7.2 (R2009b) Statistics Toolbox Software

This table summarizes what’s new in Version 7.2 (R2009b):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports

New features and changes introduced in this version are:

• “New Parallel Computing Support for Certain Functions” on pag e 6

• “New Stack a nd Unstack Methods for Dataset Arra ys ” on page 7

• “New Support for SAS Transport (.xpt) Files” on page 7

• “New Output Function in nlmefit for Monitoring or Canceling Calculations”

on page 7

Fixed Bugs an d Known Problems

Includes fixes

Related Documentation at Web Site

New Parallel Computing Support for Certain Functions

Statistics Toolbox™ now supports parallel execution for the following functions:

•

bootci

• bootstrp

• crossval

• jackknife

• TreeBagger

For more information on parallel computing in the Statistics Toolbox, see “Parallel Computing Support for Resampling Methods”.

Version 7.2 (R2009b) Statistics Toolbox™ Software

New Stack and Unstack Methods for Dataset Arrays

dataset.unstack converts a “tall” dataset array to an equivalent dataset

array that is in "wide format", by "unstacking" a single variable in the tall dataset array into multiple variables in wide. manipulation by converting a “wide” dataset array to an equivalent dataset array that is in "tall format", by "stacking up" multiple variables in the wide dataset array into a single variable in tall.

dataset.stack reverses this

New Suppor t for SAS Transport (.xpt) Files

Statistics Toolbox now supports importing and exporting files in SAS Transport (.xpt) format. For more information, see the

dataset.export reference pages.

xptread and

New Output Function in nlmefit for Monitoring or Canceling Calculations

The nlmef it function now supports using an output function to monitor or cancel calculations. For more information, see the

nlmefit reference page.

Statistics Toolbox™ Release Notes

Version 7.1 (R2009a) Statistics Toolbox Software

This table summarizes what’s new in Version 7.1 (R2009a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports

New features and changes introduced in this version are:

• “Enhanced Dataset Functionality” on page 8

• “New Naïve Bayes Class ification” on page 9

• “New Ensemble Methods for Classification and Regression Trees” on page 9

• “New Performance Curve Function” on page 9

• “New Probability Distribution Objects” on page 9

Fixed Bugs an d Known Problems

Includes fixes

Related Documentation at Web Site

Enhanced Dataset Functionality

• An enhanced dataset.join method provides additional types of join

operations:

- join can now perform m ore complicated inner and outer join operations

that allow a many-to-many correspondence between dataset arrays and B, and allow unmatched observations in either A or B.

- join can be of Type 'inne r', 'leftouter', 'rightouter', 'fullouter',

'outer' (which is a synonym for 'fullouter'). For an inner join,

the dataset array, combination of key values that occurred in both right) outer join, (or B) that did not match any in B (or A).

C, only contains observations corresponding to a

A and B. For a left (or

C also contains observations corresponding to keys in A

- join can now return index vectors indicating the correspondence

between observations in

C and those in A and B.

- join now supports using multiple keys.

Version 7.1 (R2009a) Statistics Toolbox™ Software

- join now supports an optional parameter for specifying missing key

behavior rather than raising an error.

• An enhanced

to Microsoft

dataset.export method now supports exporting directly

Excel®files.

New Naïve Bayes Classification

• The NaiveBayes classification object is suitable for data sets that contain

many predictors or features.

• It supports normal, kernel, multinomial, and multivariate multinomial

distributions.

New Ensemble Methods for Classification and Regression Trees

• New classification objects, TreeBagger and CompactTreeBagger,provide

improved performance through bootstrap aggregation (bagging).

• Includes Breiman’s “random fo rest” method.

• Enhanced

classregtree has more options for growing and pruning trees.

New Performance Curve Function

• New perfcurve function provides graphical method to evaluate

classification results.

• Includes ROC (receiver operating characteristic) and other curves.

New Probability Distribution Objects

• Provides a consistent interface for working with probability d i s trib u tions.

• Can be created directly using the

data using the

• Option to fit distributions by group.

• Includes kernel object methods and parametric object methods that you can

use to analyze the distribution represented by the object.

fitdist function.

ProbDistUnivParam constructor, or fit to

Statistics Toolbox™ Release Notes

• Includes kernel object properties and parametric object properties that you

can access to determine the fit results and evaluate their accuracy.

• Related enhancements in the

qqplot functions.

chi2gof, histfit, kstest, probplot,and

Version 7.0 (R2008b) Statistics Toolbox™ Software

Version 7.0 (R2008b) Statistics Toolbox Software

This table summarizes what’s new in Version 7.0 (R2008b):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features and changes introduced in this version are organized by these topics:

• “Classification” on page 11

• “Data Organization” on page 11

• “Model Assessment” on page 12

• “Multivariate Methods” on page 12

• “Probability Distributions” on page 12

• “Regression Analysis” on page 13

• “Statistical Visualization” on page 13

• “Utility Functions” on page 14

Fixed Bugs an d Known Problems

No No

Related Documentation at Web Site

Classification

The new confusionmat function tabulates misclassifications by comparing known and predicted classes of observatio ns.

Data Organization

Dataset arrays constructed by the dataset function can now be written to an external text file using the new

When reading ex ternal text files into a dataset array,

'TreatAsEmpty' parameter for specifying strings to be treated as empty.

export function.

dataset has a new

Statistics Toolbox™ Release Notes

Compatibility Considerations

In previous versions, dataset used eval to evaluate strings in external text files before writing them into a dataset array. As a result, strings such as

'1/1/2008' were treated as numerical expressions with two divides. Now, dataset treats such expressions as strings, and writes a string variable into

the dataset array whenever a column in the external file contains a string that does not represent a valid scalar value.

Model Assessment

The cross-validation function, crossval, has new options for directly specifying loss functions for mean-squared error or misclassification rate, without having to provide a separate function M-file.

Multivariate Methods

The procrustes function has new options for computing linear transformations without scale or reflection components.

Probability Distributions

The multivariate normal functions mvnpdf, mvncdf,andmvnrnd now accept vector specification of diagonal covariance matrices, with corresponding gains in computational efficiency.

The hypergeometric distribution has been added to both the

randtool graphical user interfaces.

Compatibility Considerations

The ksdensity function may give different answers for the case where there are censoring times beyond the last observed value. In this case,

ksdensity tries to reduce the bias in its density estimate by folding kernel

functions across a folding point so that they do not extend into the area that is completely censored. Two things have changed for this release:

1 In previous releases the folding point was the last observed value. In this

release it is the first censoring time after the last observed value.

disttool and

Version 7.0 (R2008b) Statistics Toolbox™ Software

2 The folding procedure is applied not just when the 'function' parameter

'pdf',butforall'function' values.

Regression Analysis

The new nlm efit function fits nonlinear mixed-effects models to data with both fixed and random sources of variation. Mixed-effects models are commonly used with data over multiple groups, where measurements are correlated w ithin groups but independent between groups.

Statistical Visualization

The boxplot function has new options for handling multiple grouping variables and extreme outliers.

The

lsline, gline, refline,andr efcu rve functions now work with scatter

plots produced by the worked only with scatter plots produced by the

The following visualization function s now have custom data cursors, displaying information such as observation numbers, group numbers, and the values of related variables:

scatter function. In previous versions, these functions

plot function.

•

andrewsplot

• biplot

• ecdf

• glyphplot

• gplotmatrix

• gscatter

• normplot

• parallelcoords

• probplot

• qqplot

• scatterhist

• wblplot

Statistics Toolbox™ Release Notes

Compatibility Considerations

Changes to boxplot have altered a number of default behaviors:

• Box labels are now drawn as text objects rather than tick labels. Any code

that customizes the box labels by changing tick marks should now set the tick locations as wellastheticklabels.

• The function no longer returns a handles array with a fixed number

handles, and the order and meaning of the handles now depends on which options are selected. To locate a handle of interest, search for its property using findobj. 'Tag' values for box plot components are listed on the

• There are now valid handles for outliers, even when boxes have no outliers.

In previous releases, the handles array returned by the function had values in place of handles when boxes had no outliers. Now the 'xdata' and 'y data ' for outliers are NaN when there are no outliers.

'Tag'

boxplot reference page.

NaN

• For small groups, the

'notch' parameter sometimes produces notches that

extend outside of the box. In previous releases, the notch was truncated to the extent of the box, which could produce a misleading display. A new value of

As a consequence, the

'markers' for this parameter avoids the display issue.

anova1 function, which displays notched box plots for

grouped data, may show notches that extend outside the boxes.

Utility Functions

The statistics options structure created by statset now includes a Jacobian field to specify whether or not an objective function can return the Jacobian as a second output.

Version 6.2 (R2008a) Statistics Toolbox™ Software

Version 6.2 (R2008a) Statistics Toolbox Software

This table summarizes what’s new in Version 6.2 (R2008a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features and changes introduced in this version are organized by these topics:

• “Descriptive Statistics” on page 15

• “Model Assessment” on page 16

• “Multivariate Methods” on page 16

• “Probability Distributions” on page 16

• “Regression Analysis” on page 16

• “Statistical Visualization” on page 16

• “Utility Functions” on page 16

Fixed Bugs an d Known Problems

Bug Reports Includes fixes

Related Documentation at Web Site

Descriptive Statistics

Bootstrap confidence intervals computed by bootci are n ow more accurate for lumpy data.

Compatibility Considerations

The formula for bootci confidence intervals of type 'bca' or 'cper' involves the proportion of bootstrap statistics less than the observed statistic. The formula now takes into account cases where there are many bootstrap statistics exactly equal to the observed statistic.

Statistics Toolbox™ Release Notes

Model Assessmen

Two new cross-va data and assess m applications.

Multivariate

A new sequenti subsets that

The new dimension r

nnmf f

Probabili

The new sob sets for ap designs, scramble quasi-ra

plications in Monte Carlo in tegration, space-filling experimental

and global optimization. Options allow you to skip, leap over, and

the points. The

ndom number streams for intermittent sampling.

Regress

The new p data wi

lsregress

th correlated predictors .

lidation functions,

odels in regression, classification, and clustering

Methods

al feature selection function,

optimize user-defined prediction criteria.

unction perform s nonnegative matrix factorization (NMF) for

eduction.

ty Distributions

olset

and haltonset functions produce quasi-random point

ion Analysis

function performs partial least-squares regression for

cvpartition and crossval,partition

sequentialfs, selects predictor

qrandstream function provides corresponding

Statis

The no eithe

Util

The s

TolT

par

tical Visualization

rmspec

r inside or outside specification limits.

function now shades regions of a normal density curve that are

ity Functions

tatistics options structure created by

ypeFun

ameter values, respectively.

and TolTypeX, to specify tolerances on objective functions and

statset now includes fields for

Version 6.1 (R2007b) Statistics Toolbox™ Software

Version 6.1 (R2007b) Statistics Toolbox Software

This table summarizes what’s new in Version 6.1 (R2007b):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features and changes introduced in this version are organized by these topics:

• “Cluster Analysis” on page 17

• “Design of Experiments” on page 18

• “Hypothesis Tests” on page 18

• “Probability Distributions” on page 18

• “Regression Analysis” on page 19

• “Statistical Visualization” on page 20

Fixed Bugs an d Known Problems

Bug Reports Includes fixes

Related Documentation at Web Site

Cluster Analysis

The new gmdistribution class represents Gaussian mixture distributions, where random points come from different multivariate normal distributions with certain probabilities. The models with specified means, covariances, and mixture proportions, or by fitting a mixture model with a specif ied number of components to data. Methods for the class include:

gmdistribution constructor creates mixture

•

fit — Distribution fitting function

pdf — Probability density function

•

cdf — Cumulative distribution function

•

random — Random number generator

•

cluster — Data clustering

•

Statistics Toolbox™ Release Notes

• posterior — Cluster posterior probabilities

mahal — Mahalanobis distance

•

The cutoff values, and returns a matrix of cluster assignments, with one column per cutoff value.

Compatibility Considerations

The kmeans function now returns a vector of cluster indices of length n,where n is the number of rows in the input data matrix

values. In the past, rows of X with NaN values were ignored, and the vector of cluster indices was correspondingly reduced in size. Now the vector of cluster indices contains with other toolbox functions.

Design of Experiments

A new option in the D-optimal design function c andexch specifies fixed design points in the row-exchange algorithm. A similar feature is already available for the

cluster function for hierarchical clustering now accepts a vector of

X,evenwhenX contains NaN

NaN values where rows have been ignored, consistent

daugment function, which uses the coordinate-exchange algorithm .

Hypothesis Tests

The kstest function now uses a more accurate method to calculate the p-value for a single-sample Kolmogorov-Smirnov test.

Compatibility Considerations

kstest now compares the computed p-value to the desired cutoff, rather than

comparing the test statistic to a table of values. Results may differ from those in previous releases, especially for small samples in two-sided tests where an asymptotic formula was used in the past.

Probability Distributions

A new fitting function, copulafit, has been added to the family of functions that describe dependencies among variables using copulas. The function fits parametric copulas to data, providing a link between models of marginal distributions a nd models of data correlations.

Version 6.1 (R2007b) Statistics Toolbox™ Software

A number of probability functions now have improved accuracy, especially for extreme parameter values. The functions are:

•

betainv — More accurate for probabilities in P near 1.

binocdf — More efficient and less likely to run out of mem ory for large

•

values in

binopdf — More accurate w h en the probabilities in P are on the order of

•

eps.

fcdf — More accurate when the parameter ratios V2./V1 are much less

•

than the values in

ncx2cdf — More accurate in some extreme cases that previously returned

•

poisscdf — More efficient and less likely to run out of memory for large

•

values in

tcdf — More accurate when the squares of the values in X are much less

•

than the parameters in

tinv — More accurate when the probabilities in P are very close to 0.5 and

•

the outputs are very sm all in magnitude.

Function-style syntax for

paretotails objects has been removed.

Compatibility Considerations

The changes to the probability functions listed above may lead to different, but more accurate, outputs than in previous releases.

In previous releases, syntax of the form

obj(x) for a paretotails objects obj

invoked the cdf method. This syntax now produces a warning. To e valuate the cumulative distribution function, use the syntax

cdf(obj,x).

Regression Analysis

The new corrcov function converts a covariance matrix to the corresponding correlation matrix.

The

mvregress function now supports an option to force the estimated

covariance matrix to be diagonal.

Statistics Toolbox™ Release Notes

Compatibility Considerations

In previous releases the mvregress function, when using the 'cwls' algorithm, estimated the covariance of coefficients COVB using the estimated, rather than the initial, covariance of the responses now used, and the initial and final estimates of

Statistical Visualization

The boxplot function has a new 'compact' plotstylesuitablefordisplaying large numbers of groups.

SIGMA. The initial SIGMA is

COVB differs to a degree dependent on the difference between

SIGMA.

Version 6.0 (R2007a) Statistics Toolbox™ Software

Version 6.0 (R2007a) Statistics Toolbox Software

This table summarizes what’s new in Version 6.0 (R2007a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features and changes introduced in this version are organized by these topics:

• “Data Organization” on page 21

• “Hypothesis Testing” on page 22

• “Multivariate Statistics” on page 22

• “Probability Distributions” on page 22

• “Regression Analysis” on page 23

• “Statistical Visualization” on page 24

• “Other Improvements” on page 24

Fixed Bugs an d Known Problems

Bug Reports Includes fixes

Related Documentation at Web Site

Data Organization

New categorical and dataset arrays are available for organizing and processing statistical data.

• Categorical arrays facilitate the use of nominal and ordinal categorical data.

• Dataset arrays provide a natural way to encapsulate heterogeneous

statistical data and metadata, so that it can be accessed and manipulated using familiar methods analogous to those for numerical matrices.

• Categorical and dataset arrays are supported by a variety of new functions

for manipulating the encapsulated data.

• Categorical arrays are now accepted as input arguments in all Statistics

Toolbox functions that make use of grouping variables.

Statistics Toolbox™ Release Notes

Hypothesis Test

Expanded option

• The new

parameters suc

for specifie

• The

covb outpu

suitable for u

linhyptest.

output for us

Multivaria

The new chol covarianc useful in m correlati

The

• The func

that def

• The outp

variab

e matrix, ev en if the matrix is not positive definite. Factors are

on on random number generators.

ify

class

ine b oundaries between classification regions.

s a re available for linear hypothesis testing.

linhypt

est

h as regression co efficients. These tests have the form

dvaluesof

tfrom se as the covariance matrix input argument required by The follow ing functions have been modified to return a

ewith

te Statistics

cov

function computes a Cho le sky -like decomposition of a

any of the same ways as Cholesky factors, such as imposing

function for discriminant analysis has been improved.

tion now computes the coefficients of the discriminant functions

ut of the function is now of the same type as the input grouping

group.

ing

function performs linear hypothesis tests on

H*b =

H and c,whereb is a vector of unknown parameters.

regstats and the SIGMA output from nlinfit are

covb

linhyptest: coxphfit, glmfit, mnrfit, robustfit.

Compatibility Considerations

The cl past. to a lo

1s. I

exam

Prob

New emp dis

assify

If the input argument

gical vector. In the past, output was returned as a cell array of

group is numeric, the output is now converted to the same type. For

ple, if

ability Distributions

paretotails objects are available for modeling distributions with an

irical cdf or similar distribution in the center and generalized Pareto

tributions in the tails.

function now returns outputs of different type than it did in the

group is a l ogical vector, output is now converted

group is of type uint8, the output will be of type uint8.

0sand

Version 6.0 (R2007a) Statistics Toolbox™ Software

• The paretotails function converts a data sample to a paretotails object.

The objects are useful for generating random samples from a distribution similar to the data, but with tail behavior that is less discrete than the empirical distribution.

• Objects from the

paretotails class are supported by a variety of new

methods for working with the piecewise distribution.

• The

paretotails class provides function-like behavior, so that p(x)

evaluates the cdf of p at values x.

Regression Analysis

The new mvregresslike function is a utility related to the mvregress function for fitting regression models to mu ltiv ariate data with missing values. T h e new function computes the objective (log likelihood) function, and can also compute the estimated covariance matrix for the parameter estimates.

New

classregtree objects are available for creating and analyzing

classification and regression trees.

• The

• Objects from the

• The

• The following functions now create or operate on objects from the new

classregtree function fits a classification or regression tree to

training data. The objects are useful for predicting response values from new predictors.

classregtree class are supported by a variety of new

methods for accessing information about the tree.

classregtree class provides function-like behavior, so that t(X)

evaluates the tree t at predictor values in X.

classregtree class: treefit, treedisp, treeval, treefit, treeprune, treetest.

Compatibility Considerations

Objects from the classregtree class are intended to be compatible with the structure arrays that were produced in previous ve rsions by the c l assification and regression tree functions listed above. In particular, supports dot indexing of the form t.propert y to obtain properties of the object indexing, so that

t. The class also provides function-like behavior through parenthesis

t(x) uses the tree t to classify or compute fitted values for

classregtree

Statistics Toolbox™ Release Notes

predictors x, rather than index into t as a structure array as it did in the past. As a result, cell arrays should now be used to aggregate

Statistical Visualization

The new s catt erhist function produces a scatterplot o f 2D data and illustrates the marginal distributions of the variables by drawing histograms along the two axes. The function is also useful for viewing properties of random samples produced by functions such as

lhsdesign.

Other Improvements

• The mvtrnd function now produces a single random sample from the

multivariate t distribution if the

classregtree objects.

copularnd, mvnrnd,and

cases inputargumentisabsent.

• The

zscore function, which centers and scales input data by mean and

standard deviation, now returns the means and standard deviations a s additional outputs.

Version 5.3 (R2006b) Statistics Toolbox™ Software

Version 5.3 (R2006b) Statistics Toolbox Software

This table summarizes what’s new in Version 5.3 (R2006b):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features and changes introduced in this version are organized by these topics:

• “Demos” on page 25

• “Design of Experiments” on page 25

• “Hypothesis Tests” on page 26

• “Multinomial Distribution” on page 26

• “Regression Analysis” on page 27

• “Statistical Process Control” on page 27

Fixed Bugs an d Known Problems

Bug Reports Includes fixes

Related Documentation at Web Site

Demos

The following demo has been updated:

• Selecting a Sample Size — Modified to highlight the new

function

sampsizepwr

Design of Experiments

The following visualization functions, commonly used in the design of experiments, have been added:

•

interactionplot — Two-factor interaction plot for the mean

maineffectsplot — Main effects plot for the mean

•

multivarichart — Multivari chart for the mean

•

Statistics Toolbox™ Release Notes

Hypothesis Test

The following fu

•

jbtest —Replac

which is asympt p-values from Carlo simulat

•

lillietest —U

covering a wi accurate val value distr simulation

•

runstest —A

above or be

•

sampsizep

test to hav of test ty

nctions for hypothesis testing have been added or improved:

ions to compute p-values outside of the table.

der range of sample sizes and significance levels, with more

ues. New options allow you to test for exponential and extreme ibutions, as well as normal distributions, and to run Monte Carlo stocomputep-values outside of the tables.

low a specified value.

— New function to compute the sample size necessary for a

e a specified power. Options are available for choosing a variety

pes.

es the chi-square approximation of the test statistic,

otic, with a more accurate algorithm that interpolates

a table of quantiles. A new option allows you to run Monte

ses an improved version of Lilliefors’ table of quantiles,

dds a test for runs up and down to the existing test for runs

Compatibility Considerations

If the si [0.001, previou return using t tabula

gnificance level for a test lies outside the range of tabulated values,

0.5], then both s versions,

ed an error outside a smaller range, [0.01, 0.2]. Error messages suggest

he new Monte Carlo option for computing values outside the range of

ted values.

jbtest and lillietest now return an error. In

jbtest returned an approximate p-value and lillietest

data sample for a test leads to a p-value outside the range of tabulated

If the

s, then both

value

mallest or largest tabulated value. In previous versions,

the s

proximate p-value and

an ap

Mult

The pro

•

inomial Distribution

multinomial distribution has been added to the list of almost 50

bability d istributions supported by the toolbox.

pdf

— Multinomial probability density function

jbtest and lillietest now return, with a warning, either

jbtest returned

lillietest returned NaN.

Version 5.3 (R2006b) Statistics Toolbox™ Software

• mnrnd — Multinomial random number generator

Regression Analysis

Multinomial Regression

Support has been added for multinomial regression modeling of discrete multi-category response data, including multinomial logisti c regression. The following new functions supplement the regression models in

glmval by providing for a wider range of response values:

•

mnrfit — Fits a multinomial regre ssi on model to dat a

mnrval — Computes predicted probabilities for the multinomial regression

•

model

Multivariate Regression

The new mvregress function carries out multivariate regression on data with missing response values. An option allows you to specify how missing data is handled.

glmfit and

Survival Analysis

coxphfit — A new option allows you to specify the values at which the

baseline hazard is computed.

Statistical Process Control

The following new functions consolidate and expand upon existing functions for statistical process control:

•

capability — Computes a wider range of probabilities and capability

indices than the

controlchart — Displays a wider range of control charts than the

•

ewmaplot, schart,andxbarplot functions found in previous releases

controlrules — Supplements the new controlchart function by

•

providing for a wider range of control rules (Western Electric and Nelson)

capable function found in previous releases

Statistics Toolbox™ Release Notes

• gagerr — Performs a gage repeatability and repr od ucibility study on

measurements grouped by operator and part

Compatibility Considerations

The capability function subsumes the capable function that appeared in previous versions of Statistics Toolbox software, and the function subsumes the functions ewmaplot, schart,andxbarplot.Theolder functions remain in the toolbox for backwards compatibility, but they are no longer documented or supported.

controlchart

Version 5.2 (R2006a) Statistics Toolbox™ Software

Version 5.2 (R2006a) Statistics Toolbox Software

This table summarizes what’s new in Version 5.2 (R2006a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports

New features and changes introduced in this version are organized by these topics:

• “Analysis of Variance” on page 29

• “Bootstrapping” on page 29

• “Demos” on page 30

• “Design of Experiments” on page 30

• “Hypothesis Tests” on page 30

• “Multivariate Distributions” on page 31

• “Random Number Generation” on page 31

• “Robust Regression” on page 32

• “Statistical Process Control” on page 32

Fixed Bugs an d Known Problems

Includes fixes

Related Documentation at Web Site

Analysis of Variance

Support for nested and continuous factors has been added to the anovan function for N-way analysis of variance.

Bootstrapping

To following functions have been added to supplement the existing bootstrp function for bootstrap estimation:

•

bootci — Computes confidence intervals of a bootstrapped statistic. An

option allo ws you to choose the type of the bootstrap confidence interval.

Statistics Toolbox™ Release Notes

• jackknife — Draws jackknife samples from a data set and computes

statistics on each sample

Demos

The following demos have been added to the toolbox:

• Bayesian Analysis for a Logistic Regre ssion Model

• Time Series Regression of Airline Passenger Data

The following demo has been updated to demonstrate new features:

• Random Number Generation

Design of Experiments

The new fracfactge n function finds a set of fractional factorial design generators suitable for fitting a specified model.

The following functions for D-optimal designs have been enhanced:

•

cordexch, daugment, dcovary, rowe xch — New options specify the

range of values and the number of levels for each factor, exclude factor combinations, treat factors as categorical rather than continuous, control the number of iterations, and repeat the design generation process from random starting points

•

candexch — New options control the number of iterations and repeat the

design generation process from random starting points

•

candgen — New options specify the range of values and the number of levels

for each factor, and treat factors as categorical rather than continuous

•

x2fx — New option treats factors as categorical rather than continuous

Hypothesis Tests

The new dwtest function performs a Durbin-Watson test for autocorrelation in linear regression.

Version 5.2 (R2006a) Statistics Toolbox™ Software

Multivariate Di

Two new function supplement exis thesamedistri

•

mvncdf — Cumul

distribution

•

mvtcdf — Cumul

distributio

s have been added to compute multivariate cdfs. These

ting functions for pdfs and random number generators for

butions.

ative distribution function for the multivariate normal

ative distribution function for the multivariate t

Random Numb

Copulas

New functi to model co multivari

•

copulacd

•

copulap

•

copulap

•

copular

ons have been added to the toolbox that allow you to use copulas

rrelated multivariate data and generate random numbers from

ate distributions.

— Cumulative distribution function for a copula

aram

— Copula parameters a s a function of rank correlation

— Probability density function for a copula

— Random numbers from a copula

stributions

er Generation

stat

•

copula

— Rank correlation for a copula

Markov Chain Monte Carlo Methods

llowing functions generate random numbers from nonstandard

The fo

ibutions using Markov Chain Monte Carlo methods:

distr

mple

•

mhsa

algo

slic

orithm

alg

— Generate random numbers using the Metropolis-Hasting

rithm

esample

— Generate random numbers using a slice sampling

Statistics Toolbox™ Release Notes

Pearson and Johnson Systems of Distributions

Support has been added for random number generation from Pearson and Johnson systems of distributions.

•

pearsrnd — Random numbers from a distribution in the Pearson system

johnsrnd — Random numbers from a distribution in the Johnson system

•

Robust Regression

To supplement the robustfit function, the following functions now have options for robust fitting:

•

nlinfit — Nonlinear least-squares regression

nlparci — Confidence intervals for parameters in nonlinear regression

•

nlpredci — Confidence intervals for predictions in nonlinear regression

•

Statistical Process Control

The follow ing control chart functions now support time-series objects:

•

xbarplot —Xbarplot

schart — Standard deviation chart

•

ewmaplot — Exponentially weighted moving average plot

•

Version 5.1 (R14SP3) Statistics Toolbox™ So ftware

Version 5.1 (R14SP3) Statistics Toolbox Software

This table summarizes what’s new in Version 5.1 (R14SP3):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No No No

New features and changes introduced in this version are organized by these topics:

• “Demos” on page 33

• “Descriptive Statistics” on page 34

• “Hypothesis Tests” on page 34

• “Probability Distributions” on page 35

• “Regression Analysis” on page 36

• “Statistical Visualization” on page 36

Fixed Bugs an d Known Problems

Related Documentation at Web Site

Demos

The following demos have been added to the toolbox:

• Curve Fitting and Distribution Fitting

• Fitting a Univariate Distribution Using Cumulative Probabilities

• Fitting an Orthogonal Regression Using Principal Components Analysis

• Modelling Tail Data with the Generalized Pareto Distribution

• Pitfalls in Fitting Nonlinear Models by Transfor ming to Linearity

• Weighted Nonlinear Regression

The following demo has been updated:

• Modelling Data with the Generalized E x tre me Va lu e Distribution

Statistics Toolbox™ Release Notes

Descriptive Statistics

The new partialcorr function computes the correlation of one set of variables while controlling for a second set of variables.

The for grouped data. Choices include the mean, standard error of the mean, number of elements, group name, standard deviation, variance, confidence interval for the mean, and confidence interval for new observations. The function also supports the computation of user-defined statistics.

Hypothesis Tests

Chi-Square Goodness-of-Fit Test

The new chi2gof function tests if a sample comes from a specified distribution, against the alternative that it does not come from that distribution, using a chi-square test statistic.

grpstats function now computes a wider variety of descriptive statistics

Variance Tests

Threefunctionshavebeenaddedtotestsamplevariances:

•

vartest — One-sample chi-square variance test. Tests if a sample comes

from a normal distribution with specified variance, against the alternative that it comes from a normal distribution with a different variance.

•

vartest2 —Two-sampleF-test for equal variances. Tests if two

independent samples come from normal distributions with the same variance, against the alternative that they come from normal distributions with different variances.

•

vartestn — Bartlett multiple-sample test for equal variances. Tests if

multiple samples com e from normal distributions with the same vari ance, against the alternative that they come from normal distributions with different variances.

Ansari-Bradley Test

The new ansaribradley function tests if two independent samples come from the same distribution, against the alternative that they come from distributions that have the same median and shape but different variances.

Version 5.1 (R14SP3) Statistics Toolbox™ So ftware

Tests of Randomness

The new runstest function tests if a sequence of values comes in random order, against the alternative that the ordering is not random.

Probability Distributions

Support has been added for two new distributions:

• “Generalized Extreme Value Distribution” on page 35

• “Generalized Pareto Distribution” on page 35

Generalized Extreme Value Distribution

The Generalized Extreme Value distribution combines the Gumbel, Frechet, and Weibull distributions into a single distribution. It is used to model extreme values in data.

The following distribution functions have been added:

•

gevcdf — Cumulative distribution function

gevfit — Parameter estimation function

•

gevinv — Inverse cumulative distribution function

•

gevlike — Negative log-likelihood function

•

gevpdf — P robability density function

•

gevrnd — Random number generator

•

gevstat — Distribution statistics

•

Generalized Pareto Distribution

The Generalized Pareto distribution is used to model the tails of a data distribution.

The following distribution functions have been added:

•

gpcdf — Cumulative distribution function

gpfit — Parame ter estimation function

•

Statistics Toolbox™ Release Notes

• gpinv —Inversecumulativedistribution function

gplike — Negative log-likelihood function

•

gppdf — P robability density function

•

gprnd — Random number generator

•

gpstat — Distribution statistics

•

Regression Analysis

• The new c oxph fit function fits Cox’s proportional hazards regression

model to data.

• The new

simple linear regression.

• The

interval computed.

invpred function estimates the inverse prediction intervals for

polyconf function has new options to let you specify the confidence

Statistical Visualization

Both the ecdf and ksdensity functions now produce plots when no output arguments are specified.

Version 5.0.2 (R14SP2) Statistics Toolbox™ Software

Version 5.0.2 (R14SP2) Statistics Toolbox Software

This table summarizes what’s new in Version 5.0.2 (R14SP2):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports

New features and changes introduced in this version are organized by this topic:

Fixed Bugs an d Known Problems

Includes fixes

Related Documentation at Web Site

Multivariate Statistics

The cophenet function now returns cophenetic distances as well as the cophenetic correlation coefficient.

Statistics Toolbox™ Release Notes

Compatibility Summary for Statistics Toolbox Software

This table summarizes new features and changes that might cause incompatibilities when you upgrade from an earlier version, or wh en you use files on multiple versions. Details are provided in the description of the new feature or change.

Version (Release) New Features and Changes with

Version Compatib ility Impact

Latest Version V7.3 (R2010a)

V7.2 (R2009b)

V7.1 (R2009a)

V7.0 (R2008b) See the Compatibility

V6.2 (R2008a) See the Compatibility

None

Considerations subheading for each of these new features and changes:

• “Data Organization” on page 11

• “Statistical Visualization” on page

Considerations subheading for this change:

• “Descriptive S tatistics” on page 15

Compatibility Summary for Statistics Toolbox™ Software

Version (Release) New Features and Changes with

Version Compatib ility Impact

V6.1 (R2007b) See the Compatibility

Considerations subheading

for each of these new features and changes:

• “Cluster Analysis” on page 17

• “Hypothesis Tests” on page 18

• “Probability Distributions” on

page 18

• “Regression Analysis” on page 19

V6.0 (R2007a) See the Compatibility

Considerations subheading for each of these new features and changes:

• “Multivariate Statistics” on page

• “Regression Analysis” on page 23

V5.3 (R2006b) See the Compatibility

Considerations subheading for each of these new features and changes:

• “Hypothesis Tests” on page 26

• “Statistical Process Control” on

page 27

V5.2 (R2006a)

V5.1 (R14SP3)

V5.0.2 (R14SP2)

None

Mathworks STATISTICS TOOLBOX RELEASE NOTES

Specifications and Main Features

Frequently Asked Questions

User Manual