Mehr Statistik, Mehr Graphen, Weniger Aufwand!
Contents
1. A li 1 N A y GO X 1 fi i i i f i al f Y A Li A A i ii A HIN A Jf IT j y N l l 1 j fi j wn Ny o bo n A 4 NS sue ASS So i mn di i e gt SE a ee br i b A bilio di Mehr Statistik Mehr Graphen Weniger Aufwand i WN i i i r i A 1 N i pF CERF i ag MN N NA Vi N SYSTAT 12 Mehr brauchen Sie nicht SYSTAT hat alles was Sie f r die statistische Datenanalyse die Durchf hrung von Simulationen und die Pr sentation Ihrer Ergebnisse ben tigen Begleitet von ansprechenden Graphiken und ausgestattet mit einer anwenderfreundlichen Umgebung ist SYSTAT fur den Statistik Novizen ebenso geeignet wie f r den alten Statistik Hasen der lieber mit der intuitiven Kommandozeile arbeitet WE F hren Sie umf ngliche Analysen univariater und multivariater Daten basierend auf linearen allgemeinen linearen oder linearen gemischten Modellen aus m Wahlen Sie eine robuste Regressionsanalyse wenn Ihre Daten nicht f r konventionelle mehrfache Regressionsanalysen geeignet sind m F hren Sie problemlos verschiedene Simulationen durch das breite Spektrum an univariaten und multivariaten diskreten und kontinuierlichen Verteilungen f r Berechnungen Fitting und Modellierung sowie zur Generierung von Zufallszahlen macht s m glich I Erweitern Sie Ihre Simula2. Jackknife Bootstrapped estimates bias and confidence intervals 4 E Bayesian m Prior distribution multivariate normal gamma distribution m Bayes estimates and credible intervals for regression coefficients computed m Parameters of the posterior distribution provided m Plots of prior and posterior densities of regression coefficients E Ridge m Two types of ridge coefficients standardized coefficients and unstandardized coefficients m Quick Graph A plot of the ridge factor against the ridge coefficients Partial Least Sguares Regression E NIPALS algorithm E SIMPLS algorithm WE Crossvalidation Ml Standard errors Robust Regression ml Least Median of Squares LMS regression E Rank Regression E Least Trimmed Squares LTS regression Ml M regression Ml Scaled regression Least Absolute Deviation LAD regression Logistic Regression E Binary multinomial discrete choice and conditional I AIC AlCc BIC computation E Robust standard errors E Dummy variables and interactions E Deciles of risk quantiles and simulation Wl Quick Graph ROC curve for binary logistic regression E Forward backward automatic and interactive stepwise regression Probit Regression E Dummy variables and interactions I AIC AlCc BIC computation Nonlinear Regression ml Gauss Newton Quasi Newton Simplex ml Output predicted values residuals asymptotic standard errors and correlations confidence curves and regions m Sp
3. Models HLM m Specify effects as fixed or random Ml Autocorrelated error structures ml Nested Models 2 Level Repeated Measures Clustered Data E Unbalanced or balanced data Quick Graph Scatterplot histogram or scatterplot matrix of empirical Bayes estimates Hypothesis Testing MF Resampling Bootstrap without replacement Jackknife Bl Mean One Sample z test Two sample z test One Sample t test Two Sample t test Paired t test Poisson test with Bonferroni Dunn Sidak adjustments E Variance Single Variance Equality of Two Variances Equality of Several Variances I Correlation Zero Correlation Specific Correlation Equality of Two Correlations E Proportion Single Proportion Equality of Two Proportions E Appropriate Quick Graphs E Resampling Bootstrap without replacement Jackknife One way ANOVA multiple comparison tests Bonferroni Tukey Kramer HSD Scheffe Fisher s LSD Tukey s b Students Newman Keuls Duncan R E G W Q Hochberg s GT2 Gabriel Tamhane T2 Games Howell Dunnett s T3 Sidak Two way ANOVA post hoc tests on least squares means Bonferroni Tukey LSD Scheffe E Repeated measures one way two or more factors three or more factors E Designs unbalanced randomized block complete block fractional factorial mixed model nested split plot Latin square crossover and change over Hotelling s T2 m ANCOVA m Means model for missing cells designs I AIC AlCc BIC c
4. Resampling Bootstrap without replacement Jackknife E K means and K medians m Distance measures Euclidean MWSS gamma Pearson R sguared Minkowski chi sguare phi sguare absolute Mahalanobis m Quick Graphs parallel coordinate and mean std deviati on profile plots E Additive trees m Input similarity dissimilarity matrices m Quick Graph dendrogram I Five indices for cluster validity RMSTTD Dunn Da vies Bouldin Pseudo F Pseudo T2 W Cutting cluster tree based on specified nodes and tree height Correspondence Analysis ml Simple and multiple raw data and data in tabular form Quick Graphs vector and casewise plots E Resampling Bootstrap without replacement Jackknife Classification and Regression Trees E Loss functions least squares trimmed mean LAD phi coefficient Gini index twoing ml Quick Graph unique tree mobile including split statistics and color coded subgroup densities box dot dit jitter stripe E Resampling Bootstrap without replacement Jackknife Cronbach s Alpha E Resampling Bootstrap without replacement Jackknife Test Item Analysis E Classical analysis E One and two parameter logistic model ml Quick Graph item plot Multidimensional Scaling Bl Two way scaling Kruskal Guttman Young E Three way scaling INDSCAL E Non metric unfolding E EM estimation E Power scaling for ratio data ml Quick Graphs MDS plot Shepard diagram Partially Ordered Scal
5. S E Also import export ASCII Microsoft Excel SAS SPSS ODBC dBASE and ArcView file formats E Use numeric string or data time variables W Specify date as well as time formats simultaneously for any given variable E Store variable labels comments width and format value labels file comments as well as category frequency weight and ID variable information to the data file W Global options to turn off saving category and ID variable information to the data file W View multiple data files and activate any given data file from among the open data files for editing and or analysis W Paste data as text paste variable properties paste data with custom row and column separators copy variables to the clipboard and insert them anywhere in the Data Editor E Use up to 32 000 variables unlimited cases E Field width of up to 23 with up to 14 decimal places for numeric data up to 256 characters for string data E Sort or transpose data merge or append files E Label and order categories E Manage missing values E Rank center standardize and trim variables Save data sets to temporary files that are automatically deleted on exit ml Compute new variables arithmetic operators relational operators logical operators IF THEN transformations trigonometric exponential logarithmic multivariate character date and time functions m Select cases based on a specified condition and invert case selection E Save only sel
6. W Object tracker for identifying individual objects in a graph that can then be edited using the Graph Interactivity feature WE Reposition the graph title using the mouse drag and drop Global Options Decorate your graphs with different background amp border themes Threshold limit and grid cuts for automatic hexagonal binning E Location facet eye 3 D rotation angle scale and appea rance settings for all graphs through the dialog as well as command line interface W Ability to change the image type of the graphs appearing in the Output like PNG BMP JPEG GIF or EMF Bar dot line pie profile and pyramid charts Summary Charts E Use medians instead of mean for Bar Dot Line Profile and Pyramid graphs E Bar 2 D 3 D stacked error bars repeated measures percent polar mirror mosaic E Dot 2 D 3 D line connected error bars repeated mea sures percent polar mirror E Line 2 D 3 D errors repeated measures percent mirror E Pie 2 D 3 D concentric rings offset slice E Profile 2 D 3 D stacked repeated measures percent mirror E Pyramid 2 D 3 D repeated measures percent mirror W Base line Anchor is drawn at the specified base value for bar and pyramid charts W Distinct stack and percentage options for univariate bar charts Histograms box and density plots E 2 D displays m Box plot box and whisker notched box with dot m Dot densities dit symmetric dit dot jitt
7. available E Within group and between group testing is now part of menus m MANCOVA E AIC AlCc BIC computation E Resampling Bootstrap without replacement Jackknife Factor Analysis E Principal components iterated principal axis maximum likelinood E Rotation varimax guartimax equimax orthomax oblimin Discriminant Analysis E Classical Discriminant Analysis Linear or quadratic m Prior probabilities contrasts m Output F statistics F matrix eigenvalues canonical correlations canonical scores classification matrix Wilks s lambda Lawley Hotelling Pillai and Wilks s tra ce classification tables including jackknifed canonical variables covariance and correlation matrix posterior probabilities and Mahalanobis distances m Stepwise modeling automatic forward backward and interactive stepping m Resampling Bootstrap without replacement Jackknife E Robust Discriminant Analysis m Useful when the data sets are suspected to contain outliers m Linear or guadratic analysis m Save the robust Mahalanobis distance weights and predicted group membership Cluster Analysis E Hierarchical m Distance measures Euclidean percent gamma Pearson R sguared Minkowski chi sguare phi sguare absolute Anderberg Jaccard Mahalanobis RT Russel SS m Linkage methods single complete centroid average median Ward flexible beta k neighborhood uniform weighted m Ouick Graphs dendrogram matrix and polar m
8. rigen id Grundwasser t iaa Mit Hilfe von SYSTATs r umlichem Statistik Kriging Sd J Sch tzer konnte ein Modell ber potentielle Uran Reser a i fu ven erstellt werden Die Krig Kontur auf die die tats ch mk Rin my lichen Uranmengen Werte aufgetragen wurden zeigt da sich die h chste Konzentration in dem Gebiet nahe dem L ngengrad 101 und dem Breitengrad 33 4 befindet mr gi Zn medic LL EI Se ee iL Statistik Theoretische Verteilungen Mit SYSTATs integrierten statistischen Funktionen wie z B dieser Beta Dichte vergleichen Sie Theorien mit Ex perimenten SYSTAT stellt Dichte kumulative und inverse Funktionen f r 28 Verteilungen sowie Anpassungsfunk tionen f r 25 Verteilungen zur Verf gung und bietet die M glichkeit Stichproben aus 33 Verteilungen zu ziehen Banking und Finance Risk Management SYSTATs Monte Carlo Modul wird u a zur Berechnung des Value at Risk VaR eingesetzt Hier betrachten wir ein Portfolio aus Optionen zweier korrelierter Aktien J mit dem Anfangswert 2800 Eine Stichprobe von 5000 wurde entsprechend der Verteilung der Ver nderungen im Portfolio mit Zeitintervallen von 10 Tagen und einem 10 Tage 95 VaR 511 6 simuliert Der Graph zeigt ein Histogramm der Ver nderungen im Portfolio und eine gegl ttete Version 750 250 250 750 Change in Portfolio Es ist auch moglich einen Datensatz zu betrachten wahrend ein anderer analysiert wird das spart
9. viel Zeit Egal welches Statistikpaket ein Statistiker regelm ig benutzt es lohnt sich auf jeden Fall eine Kopie von SYSTAT zu haben allein schon aus dem Grund und das ist nur einer von vielen dass es erstklassige und un bertroffene Graphiken bietet Robert T Brennan Ed D Research Associate in Sozialmedizin Harvard Universit t Medizinische Fakult t SSS SYSTAT EEE reg Weltweite B ros Nord Zentral und S damerika Europa UK und Irland Andere L nder Systat Software Inc Systat Software GmbH Systat Software Inc UK Branch Office Cranes Software International Limited 1735 Technology Drive Ste 430 Schimmelbuschstr 25 24 Vista Centre 50 Salisbury Road 4th Floor Shankar Narayana Building San Jose CA 95110 40699 Erkrath Hounslow TW4 6JQ 25 M G Road USA Germany London UK Bangalore 560001 INDIA Tel 1 800 797 7401 Fax 1 800 797 7406 E mail info usa systat com Support techsupport systat com Tel 49 0 2104 9540 Fax 49 0 2104 95410 E mail kontakt systat de Support eurotechsupport systat com E mail kontakt systat de www systat de Tel 44 0 208 538 0128 Tel 91 80 4112 0000 Fax 91 80 4123 1274 E mail asiasales systat com Support apactechsupport systat com Fax 44 0 208 538 0273 E mail info systat co uk Support eurotechsupport systat com 2007 SYSTAT Software GmbH
10. Mausklick wichtige Informationen ber Ihre Daten und k nnen Ihre Analysen beschleunigen Undo Redb erleichtert die Daten Bearbeitung Nutzen Sie die neue Undo Funktion im Daten Editor um bis zu 32 Schritte im gleichen Aus wertungsgang r ckg ngig zu machen oder rufen Sie Schritte erneut auf um z B irrt mlich gel schte Angaben wiederherzustellen u Fi TERN BEE nd BEE EM M Fw PM HM WR TY rW Y An Yn rm Nutzen Sie Themen Menus und individuell zusammengesetzte Menus und Werkzeugleisten Ersetzen Sie das vorgegebene Men durch ein Themen Men entsprechend Ihrem Anwendungs gebiet Nutzen Sie SYSTATs Flexibilit t beim Aufbau von Men s und Werkzeugleisten um diese an Ihre speziellen Arbeitsbereiche und Anforderungen anzupassen Arrangieren Sie Men s und Werk zeugleisten um schnelleren Zugang zu Funktionen zu haben die Sie h ufig nutzen Startpage Untitled AFIFLSTO www systat de Liste aller Funktionen in SYSTAT 12 Informieren Sie sich hier ber alle statistischen Verfahren Graphen und Optionen in SYSTAT 12 Soforthilfe zu allen Verfahren die Sie ben tigen bekommen Sie auch mit der Bubble Hilfe STATISTICS Descriptive Statistics E Column m Arithmetic mean median sum and number of cases m Min max range and variance L m Coefficient of variation std err of mean m Adjustable confidence intervals of mean m Skewness kurtosis including standard errors m Shapiro Wi
11. a and uncertainty coefficients E R x C ordered levels Spearman s rho Goodman Kruskal s gamma Kendall s tau b Stuart s tau c Somers D E Others Mantel Haenszel test Cochran test E Row dependent and symmetric statistics E Table of counts and percents E Cell statistics Association measures for one and two way tables along with confidence intervals W Standardized tables E Resampling Bootstrap without replacement Jackknife bootstrapped estimates bias and confidence intervals Correlations Distances and Similarities ml Continuous data Pearson correlations covariance SSCP I Distance measures Euclidean city block Bray Curtis QSK E Rank order data Spearman gamma mu2 tau b tau c Bl Unordered data phi Cramer s V contingency Goodman Kruskal s lambda uncertainty coefficients E Binomial data S2 S3 S4 S5 S6 S7 Tetrachoric Yule s O Hamman Dice Sneath Ochiai Kulczynski Gower2 WE Missing data pairwise listwise deletion EM Ml Hadi outlier detection and estimation I Probabilities Bonferroni Dunn Sidak E Quick Graph scatterplot matrix E Resampling Bootstrap without replacement Jackknife bootstrapped estimates bias and confidence intervals in the case of Pearson correlations and rank orderd data Set and Canonical Correlation E Whole semi and bipartial set correlations I Rao F R Square Shrunk R Square T Square Shrunk T Square P Square Shrunk P Square Within be
12. and tick labels number of ticks and number of pips m Modify line aesthetics like color style and thickness m Set limit lines and grid lines m Modify minimum and maximum scale for an axis m Transform the scale to the log or power scale E Element m Modify element aesthetics like line color style thick ness fill color pattern symbol type size and boundary m Change error bar settings height parameter base line bar width and label settings for summary charts m Separate a slice change to an attention map ring display slice labels and transform the scale in pie charts m Change smoothers residuals connectors partitions specify vector lines vertical spikes confidence contours and hexagonal binning for scatterplots m Set the surface type gradient and wireframe for 3 D scatterplots and function plots E Legend m Modify legend titles location layout number of rows E The Graphical User Interface is a single window with panes and tabs for displaying output data graph and command files conveniently WE Startpage to access what s new in the current version recent files interface themes and manuals get useful tips and scribble notes W Variable Editor for editing various properties of variables in the data file copy paste all or some properties of any number of variables W View value labels or data values in the Data Editor W View variable statistics and histogram for any variable on right click E W
13. ars and dialogs and link them to an interface theme E Numerous global options for each aspect of the application Ml Move the active tab to the beginning of the Viewspace Commandspace E Specify file comments in the Data Editor E Command line and dialog interface interlinked so that the hypotheses features can be conveniently accessed W Crash recovery and rescue system to retrieve any unsaved data command and output files in the unlikely event of a crash or improper shut down E Dialog boxes m Tabbed dialog boxes where various options e g Options Results and various other settings come under different tabs of the main dialog box m Drag and drop facility double click facility multiple contiguous and non contiguous selection using Ctrl and Shift Keys as well as context right click menu to ease the selection of variables in different dialog boxes m All the input fields in dialogs show tool tips with range values m cons to indicate Category variables in dialog boxes and Data Editor and freguency as well as weight variables in the Data Editor m Variable labels as tooltip on mouse hover m Keyboard shortcuts for dialog items m Grid controls for entering any number of rows of input with keyboard shortcuts for ease of use m What s this help for each item in the dialog box Data Management E Data file format with compression backward compatibility provided E import export data formats like Statview Stata Statistica JMP Minitab and
14. ch berechnen Sie die optimalen Faktor Einstellungen erstellen Sie Kontur und Desirability Plots und f hren Sie Ridge Analysen durch Nutzen Sie den K Median Algorithmus als Alternative zum K Mittelwert Eee ee TT En EEAS TES MEHR GRAPHEN Produzieren Sie attraktive Graphen schnell und bequem Nutzen Sie SYSTATs breite Palette an wissenschaftlichen und technischen Graphen um Ihre Analyse zu pr sentieren und passen Sie die Graphen nach Ihren Bed rfnissen an Der Neue Interaktive Dialog erm g licht es Ihnen mehrere Aspekte Ihrer Graphiken in einem einzigen integrierten Dialog zu ver ndern Pr sentieren Sie die Zusammenfassung gro er Datens tze ansprechend und bersichtlich Nutzen Sie SYSTATs Hexagonales Binning eine Form des bivariaten Histogramms als fortgeschrittene Version des Punktdiagramms f r gro e Datens tze Diagramme f r hexagonales Binning teilen den Bild schirm in ein hexagonales Gitter auf Der Durchmesser des Hexagons ist berall gleich w hrend die Farben von der Anzahl der x y Werte abh ngen WENIGER AUFWAND Sparen Sie Zeit und M he mit der informativen Startseite Auf der Startseite erhalten Sie direkten Zugang zu zuletzt genutzten Dateien Tipps und Themen Menus Sie haben Zugriff auf PDF Dateien der Handb cher und Sie k nnen Notizen auf den Notizblock schreiben Arbeiten Sie schneller mit der Kombination aus Kommandozeilen Oberfl che und Men Dialog Nutzen Sie das Zusammenspiel
15. ecial features Cook Weisberg confidence intervals Wald intervals Marquardting E Robust estimation absolute power trim Huber Hampel t bisquare Ramsay Andrews Tukey E Maximum likelihood estimation ml Piecewise regression kinetic models logistic model for quantal response data E Exact derivatives ml Quick Graph scatterplot with fitted curve E Resampling Bootstrap without replacement Jackknife Two Stage Least Sguares E Heteroscedasticity consistent standard errors Smooth amp Plot I 126 non parametric smoothers including LOESS E Windows fixed width or nearest neighbors E Kernels uniform Epanechnikov biweight triweight tricube Gaussian Cauchy E Method median mean polynomial robust trimmed mean m Save predicted values and residuals E Resampling Bootstrap without replacement Jackknife Mixed Model Analysis WE Variance components and linear mixed model structures Estimates of parameters by m Maximum likelihood ML m Restricted maximum likelihood REML m MIVOUE 0 in the case of variance components m ANOVA in the case of variance components WE Confidence intervals and hypothesis tests based on these estimates W Structures of covariance matrix of random effects m Variance components m Diagonal Compound symmetry m Unstructured W Structures for error matrix m Variance components Compound symmetry E AIC AlCc BIC computation Mixed Regression E Hierarchical Linear
16. ected cases or specified variables IM Recode variables instantly and conveniently option to replace or create variables with recoded values W Global option to trim leading and trailing spaces in string variable data E Matrix computations through the dialog as well as com mand line interface available for use in conjunction with other statistical features I Use BASIC control structure to manipulate data read select sort transform print save create random samples and so on I Create temporary variables and arrays I Use Mersenne Twister or Wichman Hill random number generator while generating random samples Command editor Ml Autocomplete and automatic coloring of commands E Open multiple command files W Obtain help for any phrase simply by typing it and using the context menu Command Language ml Complete coverage of menu functionality E Interactive command entry speeds analysis m Command files to automate repetitive tasks ml Command log records session history E Streamlined command syntax with informative error and warning messages wherever applicable E Create command templates with token variables Ml Commands related to Basic Statistics globally available E Insert comments at the end of a command line Ml Headers footers page setup and print preview with multiple view options E HTML based output with tabular and ASCII modes tables can be directly and conveniently copied to external appli cations without di
17. en erhalten Sie Standardfehler der gesch tzen Regressionskoeffizienten Validieren Sie die angepasste Regression durch leave one out oder willk rliche Auslassung des Kreuztabellen Verfahrens Sollte das Standardverfahren f r mehrfache lineare Regression Probleme in Ihrem Datensatz offenlegen w hlen Sie eines von SYSTATs Verfahren zur Robusten Regression LAD LTS LMS S M Reihenfolge zur L sung dieses Problems Contour plot of VISCOSITY Vs TIME TEMP Entdecken Sie schnell Gruppierungen in Daten SYSTATs Cluster Analyse gibt Ihnen eine breite Vielfalt an Distanz und hnlichkeits Matritzen Cluster Kriterien Validierungs Indizes und Schneide und Stutz Methoden um aus den Daten eine zufrieden stellende hierarchische Klassifizierung oder Gruppierung zu erhalten Nutzen Sie die neuen Verbindungs Optionen die auch uniform k Nachbarschaft flexible Beta und gewichtete Verbindung einschlie en Validieren Sie Ihre Cluster mit f nf neuen Indizes Schneiden Sie Cluster Baume basierend auf Blatt Knoten und Baum H hen Entwickeln und optimieren Sie die Proze durchf hrung Wenn Sie Daten von einem gut aufgebauten Experiment nutzen rufen Sie die Response Surface Optimization Option auf um die Faktorwerte und Proze einstellungen herauszuarbeiten mit denen die besten produktcha rakteristischen Ergebnisse erzielt werden Sch tzen Sie Response Surface Parameter f hren Sie Varianzana lyse und Signifikanz Tests dur
18. ents nations West European provinces E Eleven geographic projections WE Create map shape files Additional graphs Ml Multiplots based on Trellis plots multiple displays based on grouping variables for bar dot line profile pyramid scatterplot probability plot and quantile plot E Icon plots Chernoff faces Fourier blobs histograms profiles thermometers weather vanes stars and arrows E Parallel coordinate and Andrews Fourier plots E Function plots specify any function 2D 3D 4D Coordinates and Projection E Rectangular polar triangular and spherical coordinates ml Geographic projections gnomonic stereographic Mer cator orthographic Lambert Robinson sinusoidal Miller Peters fish eye E Advanced page view that lets you position amp resize the graphs titles amp other annotation objects for printing Interactive graphics W Single dialog box with context sensitive tabs for editing individual components of the graph Graph Frame Axis Legend and Element The changes get reflected instantly E Graph and Frame m Change background color title font coordinates and projections m Zoom resize rotate and reposition graphs frames m Change the row column matrix dimension of graph frames in multiple graphs m Change summary charts like bar dot line profile and pyramid from one type to another m Change related density types from one type to another E Axes m Control axis title font for title
19. er fuzzy and stripe m 2 D 3 D displays m Histogram counts cumulative counts control number of bars or bars widths m Normal and kernel density functions m Contour and mosaic plots m Pseudo 3 D displays mirror plots Scatterplots Quantile and Probability plots E Repeated measures contour and tiled plots E Smoothers 2 D 3 D linear quadratic DWLS step NEXPO inverse Andrews bisquare Huber and Kriging The tension parameter can be specified where applicable ml Smoother residuals E Option for limiting smoother to data range E Line connecting plot points minimum spanning tree traveling salesman path Voronoi tessellation Delaunay triangulation vectors spikes and convex hull E Size points by influence sunflower symbols E Border 2d graph for Plot Probability Plot and Ouantile Plot Other 2 D plot and SPLOM options I Hexagonal binning with desired number of cuts E 38 theoretical densities for probability plots E Smoothers log power lowess spline mean median mode midrange trimmed mean E Confidence interval contours bivariate ellipsoid bivariate centroid regression line kernel density E Display univariate densities on borders histogram box box dot dot dit frequency polygon normal Kernel fuzzy stripe jitter E High low close plots 2 D E Mirror plots 2 D E Present statistical data on maps E US states counties metro areas census tracts and related demographics E World contin
20. evor Sie tiefergehende Forschungen betreiben Dieser Auto korrelationsfunktions QuickGraph erzeugt die zyklischen Muster die Sonnenflecken zeigen Die Peaks die sich Pr oberhalb der Konfidenzbereiche kreuzen weisen auf signifikante Korrelationen ber die Zeit hin F SYSTAT 12 bietet eine Vielzahl an Standardfunktionen die kein anderes Paket ohne teure Add On Module bieten kann Zu den berzeugensten Funktionen geh ren f r mich die QuickGraphs die uns allein dadurch dass sie die Daten visuell darstellen zu besseren Statistikern machen Vielleicht die gr te Verbesserung in SYSTAT 12 ist der wirklich beein druckende Ausgabe Report Alles von H ufigkeits bis zu Faktoranalyse Ergebnissen ist sofort geeignet f r die Weiterleitung an Kollegen oder die Ver ffentlichung in Publikationen und in Farbe sieht es noch viel besser aus Epidemiologie Tuberkulose Vorf lle Mit SYSTAT k nnen Sie geographische und statistische Daten kombinieren um aussagef hige Karten zu erstellen Diese Graphik setzt Farbkodierungen zur Hervorhebung hoher Raten an Tuberkulose F llen in Europa ein Psychologie Lehrmethoden und Lernen er ean rare In diesem Experiment bei dem die Effektivit t zwei verschiedener Lehrmethoden verglichen wurde war i eine Analyse der Kovarianz ANCOVA notwendig um der Wirkung von Begabung Rechnung zu tragen Be bM Gwir Umar Level reg Geologie Schdtzung der Uran Reserven aus dem Eh
21. indow menu to view multiple tabs in the Viewspace simultaneously W Quick Access menu containing all commonly used graphi cal and statistical tools E Examples tab with one click access to all the examples in the user manual Ml Add your own examples to the Examples tab E Extensive use of drag and drop and right click mouse functionality E Faster processing speed E Fully customizable main and context right click menus set captions accelerator keys and button images Ml Simple and intuitive default menu structure option to switch to other menu structures supplied downloadable or created I Advanced customizable status bar with items to toggle global settings and data processing conditions E Record and play menu and dialog actions create new menu items with links to these E Create your own menu items linked to command files or sets of commands E Several toolbars with over 250 fully customizable tools buttons m Specify modify keyboard shortcuts ml Set menu font and animation EI Create new popup menus in the Menu Bar W Create and apply interface themes that capture the menu structure and content status bar content keyboard short cuts output scheme pane dimensions amp locations toolbar positions and content recent files and user menu items Ml Use themes supplied with the product and download additional themes from the Internet Ml Customize resources including errors and warning mes sages status bar and bubble help content toolb
22. lk normality test m N amp P Tiles Cleveland Weighted average 1 Weighted average 2 Weighted average 3 Closest Empirical CDF Empirical CDF average m Trimmed Geometric and Harmonic means m Stem and Leaf display m Resampling Bootstrap without replacement Jackknife m Bootstrapped estimates bias and confidence intervals E Row m Arithmetic mean median sum and number of cases m Min max range and variance m Coefficient of variation std err of mean m Adjustable confidence intervals of mean m Skewness kurtosis including standard errors m Shapiro Wilk normality test m N amp P Tiles Cleveland Weighted average 1 Weighted average 2 Weighted average 3 Empirical CDF Empirical CDF average Closest m Trimmed Geometric and Harmonic means m Stem and Leaf display m Resampling Bootstrap without replacement Jackknife m Bootstrapped estimates bias and confidence intervals Ml Multivariate skewness and kurtosis m Multivariate normality tests based on skewness and kurtosis Henze Zirkler test Probability Distributions ml Computes probability density function cumulative distri bution function inverse cumulative distribution function and upper tail probabilities for a wide variety of univariate discrete and continuous probability distributions ml Quick Graphs graphs of the probability density function and the cumulative distribution function for continuous distributions E Random sampling from
23. mouse m Zoom graphs in the direction of each axis or all toge ther E Graph toolbar m Selection tools for selecting a subset of plot points m Zoom in amp Zoom out feature with selection zooming or step zooming tools m Pan tool for moving drag and drop the graph within the Graph Editor m Realign multiple frames to default layout with a single click of the mouse E Annotation tools m Annotation objects like Rectangle Circle Ellipse Poly line Arrow etc can be added to the graph interactively to glorify the graphs drawn m Annotation objects can be selected repositioned and resized guite easily m Annotation properties like Line Color Style Thickness and Fill Color Pattern etc can be set and modified easily m Annotation Text can be added interactively to attach meaningful contents to graph elements which need extra attention m Annotation text font properties can be set and modified conveniently E Status bar help m Tooltips showing each individual element name for e g Frame X axis Legend Bar Scatterplot Histogram etc in the status bar on mouse hover in the Graph Editor m View properties of elements like case ID and the value against variable names for all the axes E Save charts to BMP PS EPS EMF WMF JPG PICT GIF TIFF PNG with options for setting resolution size format and color translation E Frame tracker for identifying individual frames that can then be resized and or repositioned
24. ntral composite designs Response Surface Methods E Fits a second degree polynomial to one or more responses on several factors E Output regression coefficients analysis of variance tests of significance E Optimum factor settings using canonical for each response or desirability for all responses jointly analysis EM Quick Graph Desirability plots E Contour and surface plots with fixed settings for one or more factors Power Analysis E Determine sample size to achieve a specified power E Determine power for a single sample size or a range of sample sizes ml Proportions correlations t tests z tests ANOVA one way two way generic designs E Conforms to the Hypothesis tests on means and their various options E Previously only for two sided alternatives now for one sided alternatives also BI Quick Graph power curve GRAPHICS E Use of Microsoft s 16M color palette E Flicker Free rendering of graphs in Graph Editor WE Overlay an unlimited number of graphs Automatically plot and color subgroups side by side or overlaid E Graph Gallery with a variety of graph templates W Ability to view multiple graphs facilitating comparison of two or more graphs E Interactive changing of Graph properties with support for Begin End graphs and Quick Graphs E Dynamic Explorer m Experience better dynamic control of orientation of 3D graphs with automatic rotation step by step rotation or rotation using the
25. ogram Analysis with Coordinates POSAC ml Guttman Shye algorithm automatic serialization Bi Quick Graph item plot E Resampling Bootstrap without replacement Jackknife Perceptual Mapping E MDPREF E Preference mapping vector circle ellipse E Procrustes and canonical rotations ml Quick Graph biplots Conjoint Analysis E Monotonic linear log and power WE Stress and tau loss functions ml Quick Graph utility function plot E Resampling Bootstrap without replacement Jackknife Time Series E Smoothing LOWESS moving average running median and exponential ml Seasonal adjustment E Fourier and inverse Fourier transforms E Box Jenkins ARIMA model E Specify autoregressive difference and moving average parameters E Forecast and standard errors E Polynomially distributed lags Quick Graphs series plot autocorrelation partial autocor relation cross correlation periodogram WE Trend Analysis Mann Kendall test and Sen slope estimator for nonseasonal data Ml Trend Analysis Seasonal Kendall and Homogeneity tests with Sen slope estimator Spatial Statistics E 2D amp 3D variogram Kriging and simulation E Variogram types semi covariance correlogram general relative pairwise relative semi log semimadogram E Semivariogram models spherical exponential gaussian power and hole effect E Kriging types simple ordinary nonstationary and drift E Quick Graphs variogram and contour pl
26. omputation E Automatic outlier and influential point detection Quick Graph least squares means E Resampling Bootstrap without replacement Jackknife E Type ll and Ill sums of squares WE Confidence intervals and hypothesis tests for Helmert reverse Helmert deviation and simple contrasts Ml Options to test normality and homoscedasticity assumptions General Linear Model E Any general linear model Y XB e E Any general linear hypothesis ABC D E Mixed categorical and continuous variables E Stepwise model building AIC AlCc BIC computation E Post hoc tests E Resampling Bootstrap without replacement Jackknife WE See also linear regression and ANOVA Ouality Analysis Histogram Pareto Chart Box and Whisker Plot I Process Capability Analysis E Control Charts Run Chart Shewart Control Chart Average Run Length Operating Characteristic Curve Cumulative Sum Chart Moving Average Expected Weighted Moving Average X MR Chart Regression Chart TSQ Nonparametric Tests E Independent samples Kruskal Wallis two sample Kolmo gorov Smirnov Mann Whitney E Related variables sign test Wilcoxon signed rank test Friedman test Quade test ml One sample Kolmogorov Smirnov Anderson Darling test Wald Wolfowitz runs test E Resampling Bootstrap without replacement Jackknife Ml Handles wide variety of designs Performs repeated measures analysis E Means model for missing cells designs is
27. ot E Resampling Bootstrap without replacement Jackknife Signal Detection Analysis ml Models normal Chi square exponential ml Quick Graph receiver operating characteristic curve Survival Analysis E Kaplan Meier Nelson Aalen and actuarial life tables with confidence intervals E Turnbull KM estimation EM ml Cumulative hazards and log cumulative hazards ml Cox regression parametric models exponential accelerated exponential Weibull accelerated Weibull lognormal log logistic www systat de E Type I ll and Ill censoring E Stratification time dependent covariates E Forward backward automatic and interactive stepwise regression I AIC AlCc BIC computation ml Quick Graphs survival function Cox Snell residual plot quantile reliability and hazard plots Path Analysis RAMONA I Analyze covariance or correlation matrices E MWL maximum Wishart likelihood E GLS generalized least squares ml OLS ordinary least squares ml ADFG asymptotically distribution free estimate biased Gramian E ADFU unbiased Design of experiments E Choose between Classic and Advanced DOE with dynamic wizard E Optimal Designs ml Complete and incomplete factorial designs E Latin square designs 3 12 levels per factor E Box and Hunter 2 level incomplete designs E Taguchi designs E Plackett and Burman designs E Mixture lattice centroid axial screening E Response surface designs Box Behnken and ce
28. sing in an analysis Ml Organize output based on the data file used for a given section of the output Ml Set detailed output organizer node captions E Extensive printed pdf user documentation containing 8 volume set of SYSTAT manuals Getting Started Statistics Statistics Il Statistics Ill Statistics IV Graphics Data Language Reference ml Extensive Online Help System with Index Search Favorites lists Bl Acronym expansions data file references in the Online Help system EI Knowledge base FAQ with answers to various queries raised by users WE Tutorials with step by step instructions on using various features E Status bar help and Bubble Help regarding a given feature on mouse hover on the corresponding menu item E Statistics Glossary ml Context sensitive help on pressing F1 on any item in the interface Neue Funktionen in SYSTAT 12 sind rot markiert SYSTEMANFORDERUNGEN Grundlegende Hardware und Software Anforderungen fur SYSTAT sind e Windows 32 bit Vista XP 2000 2000 Server 2003 Server e Separate Installation f r Windows 64 bit Vista XP 2003 Server Windows 98 ME NT e Pentium Rechner oder vergleichbares Level e 128 MB RAM freier Speicher 512 MB empfohlen e 220 MB freier Diskettenspeicher e CD ROM Laufwerk e SVGA Adapter und Monitor e Internet Explorer 6 oder vergleichbar www systat de SYSTAT unterst tzt alle Bereiche wissenschaftlicher und technischer Forschung Produktion Qualit
29. stortion W Collapsible links for each section of the output WE Global options m Specify the field width and number of decimal places m Short medium or long statistical output m Define font style and size for tabular as well as ASCII output formats font sizes can be condensed to be as small as desired m Display variable labels names or both m Display value labels data values or both m Wrap and or truncate text in tables appearing in the output at the desired number of characters m Control the display of statistical Quick Graphs m Global option to echo commands in the output E Page width Narrow Wide or None Infinite page width m Maximum number of characters in a row and number of columns in tables dynamically determined based on the page width and font settings m Tables with excess columns will be split into as many parts as reguired with the row and column headers appended to each part WE Global output scheme options for each component of the output settings saved to the interface theme WE Save output in the SYSTAT syo text RTF HTML or single page HTML mht formats E Output can be resumed from previous sessions using the SYSTAT format output file which contains the command log and data file information Output Organizer E Index for easy output navigation and manipulation E Combined formatted statistical output and graphs W Right click on any data file node to view it or set it active for editing or u
30. tionen und Bayes Schatzungen dann mit generischem Rejection Sampling und Markov Ketten Monte Carlo Algorithmen Ob Zeitreihen Spatiale Statistiken Mehrdimensionales Skalieren Uberlebens Testitem Cluster Korrespondenz Pfad oder Qualit tsanalyse Klassifizierungsbaume nicht parametrische Statistiken SYSTAT hat was Sie brauchen MEHR STATISTIK Analysieren Sie Komplexere Lineare Modelle Beschranken Sie sich nicht auf Analysen linearer Modelle wie Regression Varianzanalyse und Allgemei ne Lineare Modelle GLM sondern arbeiten Sie auch mit korrelierten Daten Cluster Daten abh ngigen und heteroskedastischen Daten Mit SYSTATs Analyse Gemischter Modelle analysieren Sie verschiedene Typen linearer gemischter Modelle wie Varianzkomponenten Modelle hierarchische gemischte Modelle und gemischte Regression Ob Sie verschiedene Sch tztypen fester Effekt Parameter und Varianzkomponenten oder aber Vorhersa i gen von Zufallseffekten ermitteln Konfidenz Intervalle und Hypothesen Tests geh ren dazu Em Plot of Residuals vs Predicted Values 30 8 Oom EE Bo lo o Nutzen Sie das erweiterte Spektrum der Multiplen Linearen Regression Setzen Sie die Partielle Regression der Kleinsten Quadrate ein wenn der Datensatz aus einer gro en Zahl an Vorhersage Variablen besteht gr er noch als die Anzahl an F llen SYSTAT bietet hierf r die zwei Standard Algorithmen NIPALS und SIMPLS Durch das Jackknife Verfah r
31. tsverbesserung SYSTAT unterst tzt Sie bei der berwachung Ihres Pro duktionsprozesses Hier zeigt ein X Balken QuickGraph den durchschnittlichen Widerstand von f nf zuf llig ausgesuchten Computer Komponenten gemessen ber einen Zeitraum von 20 Produktionstagen Das X Balken Diagramm zeigt da die Komponenten gut innerhalb des Kontrollbereichs liegen Ui Cher of Competes leper Arch ologie Evolution der Sch deldimensionen SYSTATS MANOVA Prozedur zeigt signifikante Varia tionen zwischen Zeitperioden in diesen Messungen an m nnlichen gyptischen Sch deln Die multi variaten Ergebnisse werden durch die univariate ANOVA die hier in diesem kombinierten Regressions Box Diagramm dargestellt ist untermauert Hecate ey Dewy ten Medizinische Forschung Klinische Versuche Die Varianzanalyse mit wiederholten Messungen ANOVA wurde hier eingesetzt um zu untersuchen wie sich vier verschiedene Wirkstoffbehandlungen j auf die H he der Bluthistamine von Hunden a auswirken ui Umweltwissenschaften TCE Kontaminierung i mg Diese Uberlebensanalyse Diagramme beschreiben J J die Wirkung vier verschiedener TCE Kontaminie E SA G EEEN rungsgrade auf Vallisneria Americana Die Diagramme lit zeigen deutlich da h here Anteile an TCE die Uber 3 Fi a n lebensraten geschlechts bergreifend reduzieren i H ah eB Jo Astronomie Sonnenfleckenzyklen Werfen Sie einen schnellen Blick auf Ihre Daten b
32. tween and inter set correlations ml Row Column betas standard errors T statistics and probabilities ml Stewart Love canonical redundancy index E Canonical coefficients loadings and redundancies E Varimax rotation E Resampling Bootstrap without replacement Jackknife Missing Value Analysis E EM Algorithm I Regression imputation E Save estimates correlation covariance SSCP matrices E Resampling Bootstrap without replacement Jackknife Loglinear Models E Full maximum likelihood E Pearson and likelihood ratio chi square E Expected values lambda SE lambda E Covariance matrix correlation matrix I Deviates Pearson deviates likelihood deviates Freeman Tukey deviates log likelihood E Resampling Bootstrap without replacement Jackknife I Dialog box with facility to type the desired model directly Linear Regression E Least squares m Cross validation saving residuals and diagnostics Durbin Watson statistic m Multiple linear regression m Prediction for new observations m Stepwise regression automatic customized and interac tive stepping partial correlations m AIC AlCc BIC computation m Hypothesis testing mixture models m Automatic outlier and influential point detection m Quick Graph residuals vs predicted values fitted model plot in the case of one or two predictors confidence and prediction intervals in the case of one predictor m Resampling Bootstrap without replacement
33. univariate and multivariate distribu tions E Fitting univariate distributions Kolmogorov Smirnov tests Anderson Darling tests Function plots Probability plots Data transforms ml Quick Graphs graphs of the respective observed and expected frequencies while fitting W 11 distributions added in SYSTAT 12 making the total number of distributions handled to 44 m Discrete Uniform Zipf Geometric Hypergeometric Ne gative Binomial Poisson Binomial Benford Logarithmic series m Continuous Triangular Double exponential Laplace Cauchy Gumbel Gompertz Lognormal Pareto Rayleigh Inverse Gaussian Lognormal Uniform Beta Normal Chi square Weibull Exponential Logistic Gamma Generalized lambda Half normal Log logistic Erlang Smallest extreme value Studentized maximum modulus Non central t Non central chi square Non central F m Multivariate distributions random sampling Multinomi al Bivariate exponential Dirichlet Multivariate normal Wishart Crosstabulation and Measures of Association E One two and multiway tables E Row and column frequencies percents expected values and deviates E List layouts order categories define intervals including missing intervals E 2 x 2 tables likelihood ratio chi square Yates Fisher s odds ratio Yule s Q E R xR tables McNemar s test Cohen s kappa E Rx C tables unordered levels phi Cramer s V contingency Goodman Kruskal s lambd
34. von Men Dialog und Kommandozeile f r eine schnellere effektivere Arbeit und entscheiden Sie selbst wie Sie die beiden Anwendungen kombinieren Sparen Sie Zeit und vermeiden Sie Fehler mit der Autocomplete Funktion SYSTATs neue Autocomplete Funktion erleichtert die Eingabe von Datei Variablen und wichtigen Kommando Namen Sobald Sie etwas eintippen gibt Ihnen SYSTAT automatisch eine Liste m glicher Namen zur Vervollst ndigung an So k nnen Sie Daten Variablen und Kommando Namen schnell und leicht wieder aufrufen und vermeiden Tipp Fehler um Ihre Arbeitsschritte zu vereinfachen Erzeugen Sie noch flexiblere Reports mit l ngeren Variablen Namen Setzen Sie Variablen Namen und Variablen Label mit bis zu 256 Buchstaben ein die K rzung der Variablen Namen im Report ist dann nicht mehr n tig Exportieren Sie Ihren Report in passender Form ohne aufw ndiges Editieren Beschleunigen Sie Ihre Analysen durch Daten und Variablen Tabs Nutzen Sie bei Ihrer Analyse den Data und den Variablen Tab im Datei Editor Der Data Tab zeigt die Daten im Arbeitsblatt an der Variablen Tab zeigt in zwei Fenstern a Informationen ber die Variablen wie Typ Label Feldweite Dezimalstellen etc b Angaben zu deren Ver arbeitung wie Gruppierungs und Gewichtungsvariablen Fall Auswahl etc Kommentare zu den Variablen k nnen Sie in einem Tooltip ablesen wenn Sie den Cursor im Data Tab auf den Variablennamen bewegen Damit erhalten Sie mit einem
Download Pdf Manuals
Related Search