### Glossary Index

###### 2

- 2D Bar/Column Plots
- 2D Box Plots
- 2D Box Plots - Box Whiskers
- 2D Box Plots - Boxes
- 2D Box Plots - Columns
- 2D Box Plots - Error Bars
- 2D Box Plots - Whiskers
- 2D Categorized Detrended Probability Plots
- 2D Categorized Half-Norm. Probability Plots
- 2D Categorized Normal Probability Plots
- 2D Detrended Probability Plots
- 2D Histograms
- 2D Histograms - Hanging Bars
- 2D Histograms - Double-Y
- 2D Line Plots
- 2D Line Plots - Aggregated
- 2D Line Plots - Double-Y
- 2D Line Plots - Multiple
- 2D Line Plots - Regular
- 2D Line Plots - XY Trace
- 2D Range Plots - Error Bars
- 2D Matrix Plots
- 2D Matrix Plots - Columns
- 2D Matrix Plots - Lines
- 2D Matrix Plots - Scatterplot
- 2D Normal Probability Plots
- 2D Probability-Probability Plots
- 2D Probability-Probability Plots-Categorized
- 2D Quantile-Quantile Plots
- 2D Quantile-Quantile Plots - Categorized
- 2D Scatterplot
- 2D Scatterplot - Categorized Ternary Graph
- 2D Scatterplot - Double-Y
- 2D Scatterplot - Frequency
- 2D Scatterplot - Multiple
- 2D Scatterplot - Regular
- 2D Scatterplot - Voronoi
- 2D Sequential/Stacked Plots
- 2D Sequential/Stacked Plots - Area
- 2D Sequential/Stacked Plots - Column
- 2D Sequential/Stacked Plots - Lines
- 2D Sequential/Stacked Plots - Mixed Line
- 2D Sequential/Stacked Plots - Mixed Step
- 2D Sequential/Stacked Plots - Step
- 2D Sequential/Stacked Plots - Step Area
- 2D Ternary Plots - Scatterplot

###### 3

- 3D Bivariate Histogram
- 3D Box Plots
- 3D Box Plots - Border-style Ranges
- 3D Box Plots - Double Ribbon Ranges
- 3D Box Plots - Error Bars
- 3D Box Plots - Flying Blocks
- 3D Box Plots - Flying Boxes
- 3D Box Plots - Points
- 3D Categorized Plots - Contour Plot
- 3D Categorized Plots - Deviation Plot
- 3D Categorized Plots - Scatterplot
- 3D Categorized Plots - Space Plot
- 3D Categorized Plots - Spectral Plot
- 3D Categorized Plots - Surface Plot
- 3D Deviation Plots
- 3D Range Plot - Error Bars
- 3D Raw Data Plots - Contour/Discrete
- 3D Scatterplots
- 3D Scatterplots - Ternary Graph
- 3D Space Plots
- 3D Ternary Plots
- 3D Ternary Plots - Categorized Scatterplot
- 3D Ternary Plots - Categorized Space
- 3D Ternary Plots - Categorized Surface
- 3D Ternary Plots - Categorized Trace
- 3D Ternary Plots - Contour/Areas
- 3D Ternary Plots - Contour/Lines
- 3D Ternary Plots - Deviation
- 3D Ternary Plots - Space
- 3D Trace Plots

###### A

- Aberration, Minimum
- Abrupt Permanent Impact
- Abrupt Temporary Impact
- Accept-Support Testing
- Accept Threshold
- Activation Function (in Neural Networks)
- Additive Models
- Additive Season, Damped Trend
- Additive Season, Exponential Trend
- Additive Season, Linear Trend
- Additive Season, No Trend
- Adjusted means
- Aggregation
- AID
- Akaike Information Criterion (AIC)
- Algorithm
- Alpha
- Anderson-Darling Test
- ANOVA
- Append a Network
- Append Cases and/or Variables
- Application Programming Interface (API)
- Arrow
- Assignable Causes and Actions
- Association Rules
- Asymmetrical Distribution
- AT&T Runs Rules
- Attribute (attribute variable)
- Augmented Product Moment Matrix
- Autoassociative Network
- Automatic Network Designer

###### B

- B Coefficients
- Back Propagation
- Bagging (Voting, Averaging)
- Balanced ANOVA Design
- Banner Tables
- Bar/Column Plots, 2D
- Bar Dev Plot
- Bar Left Y Plot
- Bar Right Y Plot
- Bar Top Plot
- Bar X Plot
- Bartlett Window
- Basis Functions
- Batch algorithms in
*STATISTICA Neural Net* - Bayesian Information Criterion (BIC)
- Bayesian Networks
- Bayesian Statistics
- Bernoulli Distribution
- Best Network Retention
- Best Subset Regression
- Beta Coefficients
- Beta Distribution
- Bimodal Distribution
- Binomial Distribution
- Bivariate Normal Distribution
- Blocking
- Bonferroni Adjustment
- Bonferroni Test
- Boosting
- Boundary Case
- Box Plot/Medians (Block Stats Graphs)
- Box Plot/Means (Block Stats Graphs)
- Box Plots, 2D
- Box Plots, 2D - Box Whiskers
- Box Plots, 2D - Boxes
- Box Plots, 2D - Whiskers
- Box Plots, 3D
- Box Plots, 3D - Border-Style Ranges
- Box Plots, 3D - Double Ribbon Ranges
- Box Plots, 3D - Error Bars
- Box Plots, 3D - Flying Blocks
- Box Plots, 3D - Flying Boxes
- Box Plots, 3D - Points
- Box-Ljung Q Statistic
- Breakdowns
- Breaking Down (Categorizing)
- Brown-Forsythe Homogeneity of Variances
- Brushing
- Burt Table

###### C

- Canonical Correlation
- Cartesian Coordinates
- Casewise Missing Data Deletion
- Categorical Dependent Variable
- Categorical Predictor
- Categorized Graphs
- Categorized Plots, 2D-Detrended Prob. Plots
- Categorized Plots, 2D-Half-Normal Prob. Plots
- Categorized Plots, 2D - Normal Prob. Plots
- Categorized Plots, 2D - Prob.-Prob. Plots
- Categorized Plots, 2D - Quantile Plots
- Categorized Plots, 3D - Contour Plot
- Categorized Plots, 3D - Deviation Plot
- Categorized Plots, 3D - Scatterplot
- Categorized Plots, 3D - Space Plot
- Categorized Plots, 3D - Spectral Plot
- Categorized Plots, 3D - Surface Plot
- Categorized 3D Scatterplot (Ternary graph)
- Categorized Contour/Areas (Ternary graph)
- Categorized Contour/Lines (Ternary graph)
- Categorizing
- Cauchy Distribution
- Cause-and-Effect Diagram
- Censoring (Censored Observations)
- Censoring, Left
- Censoring, Multiple
- Censoring, Right
- Censoring, Single
- Censoring, Type I
- Censoring, Type II
- CHAID
- Characteristic Life
- Chernoff Faces (Icon Plots)
*Chi*-square Distribution- Circumplex
- City-Block (Manhattan) Distance
- Classification
- Classification (in Neural Networks)
- Classification and Regression Trees
- Classification by Labeled Exemplars (in NN)
- Classification Statistics (in Neural Networks)
- Classification Thresholds (in Neural Networks)
- Classification Trees
- Class Labeling (in Neural Networks)
- Cluster Analysis
- Cluster Diagram (in Neural Networks)
- Cluster Networks (in Neural Networks)
- Coarse Coding
- Codes
- Coding Variable
- Coefficient of Determination
- Coefficient of Variation
- Column Sequential/Stacked Plot
- Columns (Box Plot)
- Columns (Icon Plot)
- Common Causes
- Communality
- Complex Numbers
- Conditional Probability
- Conditioning (Categorizing)
- Confidence Interval
- Confidence Interval for the Mean
- Confidence Interval vs. Prediction Interval
- Confidence Limits
- Confidence Value (Association Rules)
- Confusion Matrix (in Neural Networks)
- Conjugate Gradient Descent (in Neural Net)
- Continuous Dependent Variable
- Contour/Discrete Raw Data Plot
- Contour Plot
- Control, Quality
- Cook's Distance
- Correlation
- Correlation, Intraclass
- Correlation (Pearson r)
- Correlation Value (Association Rules)
- Correspondence Analysis
- Cox-Snell Gen. Coefficient Determination
- Cpk, Cp, Cr
- CRISP
- Cross Entropy (in Neural Networks)
- Cross Verification (in Neural Networks)
- Cross-Validation
- Crossed Factors
- Crosstabulations
- C-SVM Classification
- Cubic Spline Smoother
- "Curse" of Dimensionality

###### D

- Daniell (or Equal Weight) Window
- Data Mining
- Data Preparation Phase
- Data Reduction
- Data Rotation (in 3D space)
- Data Warehousing
- Decision Trees
- Degrees of Freedom
- Deleted Residual
- Denominator Synthesis
- Dependent t-test
- Dependent vs. Independent Variables
- Deployment
- Derivative-Free Funct. Min. Algorithms
- Design, Experimental
- Design Matrix
- Desirability Profiles
- Detrended Probability Plots
- Deviance
- Deviance Residuals
- Deviation
- Deviation Assign. Algorithms (in Neural Net)
- Deviation Plot (Ternary Graph)
- Deviation Plots, 3D
- DFFITS
- DIEHARD Suite of Tests & Randm. Num. Gen.
- Differencing (in Time Series)
- Dimensionality Reduction
- Discrepancy Function
- Discriminant Function Analysis
- Distribution Function
- DOE
- Document Frequency
- Double-Y Histograms
- Double-Y Line Plots
- Double-Y Scatterplot
- Drill-Down Analysis
- Drilling-down (Categorizing)
- Duncan's test
- Dunnett's test
- DV

###### E

- Effective Hypothesis Decomposition
- Efficient Score Statistic
- Eigenvalues
- Ellipse, Prediction Area and Range
- EM Clustering
- Endogenous Variable
- Ensembles (in Neural Networks)
- Enterprise Resource Planning (ERP)
- Enterprise SPC
- Enterprise-Wide Software Systems
- Entropy
- Epoch in (Neural Networks)
- Eps
- EPSEM Samples
- ERP
- Error Bars (2D Box Plots)
- Error Bars (2D Range Plots)
- Error Bars (3D Box Plots)
- Error Bars (3D Range Plots)
- Error Function (in Neural Networks)
- Estimable Functions
- Euclidean Distance
- Euler's e
- Exogenous Variable
- Experimental Design
- Explained Variance
- Exploratory Data Analysis
- Exponential Distribution
- Exponential Family of Distributions
- Exponential Function
- Exponentially Weighted Moving Avg. Line
- Extrapolation
- Extreme Values (in Box Plots)
- Extreme Value Distribution

###### F

- F Distribution
- FACT
- Factor Analysis
- Fast Analysis Shared Multidimensional Info. FASMI
- Feature Extraction (vs. Feature Selection)
- Feature Selection
- Feedforward Networks
- Fisher LSD
- Fixed Effects (in ANOVA)
- Free Parameter
- Frequencies, Marginal
- Frequency Scatterplot
- Frequency Tables
- Function Minimization Algorithms

###### G

- g2 Inverse
- Gains Chart
- Gamma Coefficient
- Gamma Distribution
- Gaussian Distribution
- Gauss-Newton Method
- General ANOVA/MANOVA
- General Linear Model
- Generalization (in Neural Networks)
- Generalized Additive Models
- Generalized Inverse
- Generalized Linear Model
- Genetic Algorithm
- Genetic Algorithm Input Selection
- Geometric Distribution
- Geometric Mean
- Gibbs Sampler
- Gini Measure of Node Impurity
- Gompertz Distribution
- Goodness of Fit
- Gradient
- Gradient Descent
- Gradual Permanent Impact
- Group Charts
- Grouping (Categorizing)
- Grouping Variable
- Groupware

###### H

- Half-Normal Probability Plots
- Half-Normal Probability Plots - Categorized
- Hamming Window
- Hanging Bars Histogram
- Harmonic Mean
- Hazard
- Hazard Rate
- Heuristic
- Heywood Case
- Hidden Layers (in Neural Networks)
- High-Low Close
- Histograms, 2D
- Histograms, 2D - Double-Y
- Histograms, 2D - Hanging Bars
- Histograms, 2D - Multiple
- Histograms, 2D - Regular
- Histograms, 3D Bivariate
- Histograms, 3D - Box Plots
- Histograms, 3D - Contour/Discrete
- Histograms, 3D - Contour Plot
- Histograms, 3D - Spikes
- Histograms, 3D - Surface Plot
- Hollander-Proschan Test
- Hooke-Jeeves Pattern Moves
- Hosmer-Lemeshow Test
- HTM
- HTML
- Hyperbolic Tangent (tanh)
- Hyperplane
- Hypersphere

###### I

- Icon Plots
- Icon Plots - Chernoff Faces
- Icon Plots - Columns
- Icon Plots - Lines
- Icon Plots - Pies
- Icon Plots - Polygons
- Icon Plots - Profiles
- Icon Plots - Stars
- Icon Plots - Sun Rays
- Increment vs Non-Increment Learning Algr.
- Independent Events
- Independent t-test
- Independent vs. Dependent Variables
- Industrial Experimental Design
- Inertia
- Inlier
- In-Place Database Processing (IDP)
- Interactions
- Interpolation
- Interval Scale
- Intraclass Correlation Coefficient
- Invariance Const. Scale Factor ICSF
- Invariance Under Change of Scale (ICS)
- Inverse Document Frequency
- Ishikawa Chart
- Isotropic Deviation Assignment
- Item and Reliability Analysis
- IV

###### J

###### K

###### L

- Lack of Fit
- Lambda Prime
- Laplace Distribution
- Latent Semantic Indexing
- Latent Variable
- Layered Compression
- Learned Vector Quantization (in Neural Net)
- Learning Rate (in Neural Networks)
- Least Squares (2D graphs)
- Least Squares (3D graphs)
- Least Squares Estimator
- Least Squares Means
- Left and Right Censoring
- Levenberg-Marquardt Algorithm (in Neural Net)
- Levene's Test for Homogeneity of Variances
- Leverage values
- Life Table
- Life, Characteristic
- Lift Charts
- Likelihood
- Lilliefors test
- Line Plots, 2D
- Line Plots, 2D - Aggregated
- Line Plots, 2D (Case Profiles)
- Line Plots, 2D - Double-Y
- Line Plots, 2D - Multiple
- Line Plots, 2D - Regular
- Line Plots, 2D - XY Trace
- Linear (2D graphs)
- Linear (3D graphs)
- Linear Activation function
- Linear Modeling
- Linear Units
- Lines (Icon Plot)
- Lines (Matrix Plot)
- Lines Sequential/Stacked Plot
- Link Function
- Local Minima
- Locally Weighted (Robust) Regression
- Logarithmic Function
- Logistic Distribution
- Logistic Function
- Logit Regression and Transformation
- Log-Linear Analysis
- Log-Normal Distribution
- Lookahead (in Neural Networks)
- Loss Function
- LOWESS Smoothing

###### M

- Machine Learning
- Mahalanobis Distance
- Mallow's CP
- Manifest Variable
- Mann-Scheuer-Fertig Test
- MANOVA
- Marginal Frequencies
- Markov Chain Monte Carlo (MCMC)
- Mass
- Matching Moments Method
- Matrix Collinearity
- Matrix Ill-Conditioning
- Matrix Inverse
- Matrix Plots
- Matrix Plots - Columns
- Matrix Plots - Lines
- Matrix Plots - Scatterplot
- Matrix Rank
- Matrix Singularity
- Maximum Likelihood Loss Function
- Maximum Likelihood Method
- Maximum Unconfounding
- MD (Missing data)
- Mean
- Mean/S.D. Algorithm (in Neural Networks)
- Mean, Geometric
- Mean, Harmonic
- Mean Substitution of Missing Data
- Means, Adjusted
- Means, Unweighted
- Median
- Meta-Learning
- Method of Matching Moments
- Minimax
- Minimum Aberration
- Mining, Data
- Missing values
- Mixed Line Sequential/Stacked Plot
- Mixed Step Sequential/Stacked Plot
- Mode
- Model Profiles (in Neural Networks)
- Models for Data Mining
- Monte Carlo
- Multi-Pattern Bar
- Multicollinearity
- Multidimensional Scaling
- Multilayer Perceptrons
- Multimodal Distribution
- Multinomial Distribution
- Multinomial Logit and Probit Regression
- Multiple Axes in Graphs
- Multiple Censoring
- Multiple Dichotomies
- Multiple Histogram
- Multiple Line Plots
- Multiple Scatterplot
- Multiple R
- Multiple Regression
- Multiple Response Variables
- Multiple-Response Tables
- Multiple Stream Group Charts
- Multiplicative Season, Damped Trend
- Multiplicative Season, Exponential Trend
- Multiplicative Season, Linear Trend
- Multiplicative Season, No Trend
- Multivar. Adapt. Regres. Splines MARSplines
- Multi-way Tables

###### N

- Nagelkerke Gen. Coefficient Determination
- Naive Bayes
- Neat Scaling of Intervals
- Negative Correlation
- Negative Exponential (2D graphs)
- Negative Exponential (3D graphs)
- Neighborhood (in Neural Networks)
- Nested Factors
- Nested Sequence of Models
- Neural Networks
- Neuron
- Newman-Keuls Test
- N-in-One Encoding
- Noise Addition (in Neural Networks)
- Nominal Scale
- Nominal Variables
- Nonlinear Estimation
- Nonparametrics
- Non-Outlier Range
- Nonseasonal, Damped Trend
- Nonseasonal, Exponential Trend
- Nonseasonal, Linear Trend
- Nonseasonal, No Trend
- Normal Distribution
- Normal Distribution, Bivariate
- Normal Fit
- Normality Tests
- Normalization
- Normal Probability Plots
- Normal Probability Plots (Computation Note)
- n Point Moving Average Line

###### O

- ODBC
- Odds Ratio
- OLE DB
- On-Line Analytic Processing (OLAP)
- One-Off (in Neural Networks)
- One-of-N Encoding (in Neural Networks)
- One-Sample t-Test
- One-Sided Ranges Error Bars Range Plots
- One-Way Tables
- Operating Characteristic Curves
- Ordinal Multinomial Distribution
- Ordinal Scale
- Outer Arrays
- Outliers
- Outliers (in Box Plots)
- Overdispersion
- Overfitting
- Overlearning (in Neural Networks)
- Overparameterized Model

###### P

- Pairwise Del. Missing Data vs Mean Subst.
- Pairwise MD Deletion
- Parametric Curve
- Pareto Chart Analysis
- Pareto Distribution
- Part Correlation
- Partial Correlation
- Partial Least Squares Regression
- Partial Residuals
- Parzen Window
- Pearson Correlation
- Pearson Curves
- Pearson Residuals
- Penalty Functions
- Percentiles
- Perceptrons (in Neural Networks)
- Pie Chart
- Pie Chart - Counts
- Pie Chart - Multi-Pattern Bar
- Pie Chart - Values
- Pies (Icon Plots)
- PMML (Predictive Model Markup Language)
- PNG Files
- Poisson Distribution
- Polar Coordinates
- Polygons (Icon Plots)
- Polynomial
- Population Stability Report
- Portable Network Graphics Files
- Positive Correlation
- Post hoc Comparisons
- Post Synaptic Potential (PSP) Function
- Posterior Probability
- Power (Statistical)
- Power Goal
- Ppk, Pp, Pr
- Prediction Interval Ellipse
- Prediction Profiles
- Predictive Data Mining
- Predictive Mapping
- Predictive Model Markup Language (PMML)
- Predictors
- PRESS Statistic
- Principal Components Analysis
- Prior Probabilities
- Probability
- Probability Plots - Detrended
- Probability Plots - Normal
- Probability Plots - Half-Normal
- Probability-Probability Plots
- Probability-Probability Plots - Categorized
- Probability Sampling
- Probit Regression and Transformation
- PROCEED
- Process Analysis
- Process Capability Indices
- Process Performance Indices
- Profiles, Desirability
- Profiles, Prediction
- Profiles (Icon Plots)
- Pruning (in Classification Trees)
- Pseudo-Components
- Pseudo-Inverse Algorithm
- Pseudo-Inverse-Singular Val. Decomp. NN
- PSP (Post Synaptic Potential) Function
- Pure Error
- p-Value (Statistical Significance)

###### Q

###### R

- R Programming Language
- Radial Basis Functions
- Radial Sampling (in Neural Networks)
- Random Effects (in Mixed Model ANOVA)
- Random Forests
- Random Num. from Arbitrary Distributions
- Random Numbers (Uniform)
- Random Sub-Sampling in Data Mining
- Range Ellipse
- Range Plots - Boxes
- Range Plots - Columns
- Range Plots - Whiskers
- Rank
- Rank Correlation
- Ratio Scale
- Raw Data, 3D Scatterplot
- Raw Data Plots, 3D - Contour/Discrete
- Raw Data Plots, 3D - Spikes
- Raw Data Plots, 3D - Surface Plot
- Rayleigh Distribution
- Receiver Oper. Characteristic Curve
- Receiver Oper. Characteristic (in Neural Net)
- Rectangular Distribution
- Regression
- Regression (in Neural Networks)
- Regression, Multiple
- Regression Summary Statistics (in Neural Net)
- Regular Histogram
- Regular Line Plots
- Regular Scatterplot
- Regularization (in Neural Networks)
- Reject Inference
- Reject Threshold
- Relative Function Change Criterion
- Reliability
- Reliability and Item Analysis
- Representative Sample
- Resampling (in Neural Networks)
- Residual
- Resolution
- Response Surface
- Right Censoring
- RMS (Root Mean Squared) Error
- Robust Locally Weighted Regression
- ROC Curve
- ROC Curve (in Neural Networks)
- Root Cause Analysis
- Root Mean Square Stand. Effect RMSSE
- Rosenbrock Pattern Search
- Rotating Coordinates, Method of
- r (Pearson Correlation Coefficient)
- Runs Tests (in Quality Control)

###### S

- Sampling Fraction
- Scalable Software Systems
- Scaling
- Scatterplot, 2D
- Scatterplot, 2D-Categorized Ternary Graph
- Scatterplot, 2D - Double-Y
- Scatterplot, 2D - Frequency
- Scatterplot, 2D - Multiple
- Scatterplot, 2D - Regular
- Scatterplot, 2D - Voronoi
- Scatterplot, 3D
- Scatterplot, 3D - Raw Data
- Scatterplot, 3D - Ternary Graph
- Scatterplot Smoothers
- Scheffe's Test
- Score Statistic
- Scree Plot, Scree Test
- S.D. Ratio
- Semi-Partial Correlation
- SEMMA
- Sensitivity Analysis (in Neural Networks)
- Sequential Contour Plot, 3D
- Sequential/Stacked Plots, 2D
- Sequential/Stacked Plots, 2D - Area
- Sequential/Stacked Plots, 2D - Column
- Sequential/Stacked Plots, 2D - Lines
- Sequential/Stacked Plots, 2D - Mixed Line
- Sequential/Stacked Plots, 2D - Mixed Step
- Sequential/Stacked Plots, 2D - Step
- Sequential/Stacked Plots, 2D - Step Area
- Sequential Surface Plot, 3D
- Sets of Samples in Quality Control Charts
- Shapiro-Wilks' W test
- Shewhart Control Charts
- Short Run Control Charts
- Shuffle, Back Propagation (in Neural Net)
- Shuffle Data (in Neural Networks)
- Sigma Restricted Model
- Sigmoid Function
- Signal Detection Theory
- Simple Random Sampling (SRS)
- Simplex Algorithm
- Single and Multiple Censoring
- Singular Value Decomposition
- Six Sigma (DMAIC)
- Six Sigma Process
- Skewness
- Slicing (Categorizing)
- Smoothing
- SOFMs Self-Organizing Maps Kohonen Net
- Softmax
- Space Plots 3D
- SPC
- Spearman R
- Special Causes
- Spectral Plot
- Spikes (3D graphs)
- Spinning Data (in 3D space)
- Spline (2D graphs)
- Spline (3D graphs)
- Split Selection (for Classification Trees)
- Splitting (Categorizing)
- Spurious Correlations
- SQL
- Square Root of the Signal to Noise Ratio (f)
- Stacked Generalization
- Stacking (Stacked Generalization)
- Standard Deviation
- Standard Error
- Standard Error of the Mean
- Standard Error of the Proportion
- Standardization
- Standardized DFFITS
- Standardized Effect (Es)
- Standard Residual Value
- Stars (Icon Plots)
- Stationary Series (in Time Series)
- STATISTICA Advanced Linear/Nonlinear
- STATISTICA Automated Neural Networks
- STATISTICA Base
- STATISTICA Data Miner
- STATISTICA Data Warehouse
- STATISTICA Document Management System
- STATISTICA Enterprise
- STATISTICA Enterprise/QC
- STATISTICA Enterprise Server
- STATISTICA Enterprise SPC
- STATISTICA Monitoring and Alerting Server
- STATISTICA MultiStream
- STATISTICA Multivariate Stat. Process Ctrl
- STATISTICA PI Connector
- STATISTICA PowerSolutions
- STATISTICA Process Optimization
- STATISTICA Quality Control Charts
- STATISTICA Sequence Assoc. Link Analysis
- STATISTICA Text Miner
- STATISTICA Variance Estimation Precision
- Statistical Power
- Statistical Process Control (SPC)
- Statistical Significance (p-value)
- Steepest Descent Iterations
- Stemming
- Steps
- Stepwise Regression
- Stiffness Parameter (in Fitting Options)
- Stopping Conditions
- Stopping Conditions (in Neural Networks)
- Stopping Rule (in Classification Trees)
- Stratified Random Sampling
- Stub and Banner Tables
- Studentized Deleted Residuals
- Studentized Residuals
- Student's t Distribution
- Sum-Squared Error Function
- Sums of Squares (Type I, II, III (IV, V, VI))
- Sun Rays (Icon Plots)
- Supervised Learning (in Neural Networks)
- Support Value (Association Rules)
- Support Vector
- Support Vector Machine (SVM)
- Suppressor Variable
- Surface Plot (from Raw Data)
- Survival Analysis
- Survivorship Function
- Sweeping
- Symmetrical Distribution
- Symmetric Matrix
- Synaptic Functions (in Neural Networks)

###### T

- Tables
- Tapering
- t Distribution (Student's)
- Tau, Kendall
- Ternary Plots, 2D - Scatterplot
- Ternary Plots, 3D
- Ternary Plots, 3D - Categorized Scatterplot
- Ternary Plots, 3D - Categorized Space
- Ternary Plots, 3D - Categorized Surface
- Ternary Plots, 3D - Categorized Trace
- Ternary Plots, 3D - Contour/Areas
- Ternary Plots, 3D - Contour/Lines
- Ternary Plots, 3D - Deviation
- Ternary Plots, 3D - Space
- Text Mining
- THAID
- Threshold
- Time Series
- Time Series (in Neural Networks)
- Time-Dependent Covariates
- Tolerance (in Multiple Regression)
- Topological Map
- Trace Plots, 3D
- Trace Plot, Categorized (Ternary Graph)
- Training/Test Error/Classification Accuracy
- Transformation (Probit Regression)
- Trellis Graphs
- Trimmed Means
- t-Test (independent & dependent samples)
- Tukey HSD
- Tukey Window
- Two-State (in Neural Networks)
- Type I, II, III (IV, V, VI) Sums of Squares
- Type I Censoring
- Type II Censoring
- Type I Error Rate

###### U

###### V

###### W

###### X

###### Y

###### Z

Daniell (or Equal Weight) Window. In Time Series, the Daniell window (Daniell 1946) is a weighted moving average transformation used to smooth the periodogram values. This transformation amounts to a simple (equal weight) moving average transformation of the periodogram values, that is, each spectral density estimate is computed as the mean of the m/2 preceding and subsequent periodogram values.

See also, Basic Notations and Principles.

Data Mining. StatSoft defines *data mining* as an analytic process designed to explore large amounts of (typically business or market related) data in search for consistent patterns and/or systematic relationships between variables, and then to validate the findings by applying the detected patterns to new subsets of data. *Data mining* uses many of the principles and techniques traditionally referred to as *Exploratory Data Analysis (EDA)*. For more information, see Data Mining.

Data Preparation Phase. In Data Mining, the input data are often "noisy," containing many errors, and sometimes information in unstructured form (e.g., in Text Mining). For example, suppose you wanted to analyze a large database of information collected on-line via the web, based on voluntary responses of persons reviewing your web site (e.g., potential customers of a web-based retailer, who filled out suggestion forms). In those instances it is very important to first verify and "clean" the data in a data preparation phase, before applying any analytic procedures. For example, some individuals might enter clearly faulty information (e.g., age = 300), either by mistake or intentionally. If those types of data errors are not detected prior to the analysis phase of the data mining project, they can greatly bias the result, and potentially cause unjustified conclusions. Typically, during the data preparation phase, the data analyst applies "filters" to the data, to verify correct data ranges, and to delete impossible co-occurrences of values (e.g., Age=5; Retired=Yes).

Data Reduction. The term *Data Reduction* is used in two distinctively different meanings:

**Data Reduction by decreasing the dimensionality (exploratory multivariate statistics).** This interpretation of the term *Data Reduction* pertains to analytic methods (typically *multivariate exploratory techniques* such as Factor Analysis, Multidimensional Scaling, Cluster Analysis, Canonical Correlation, or Neural Networks) that involve reducing the dimensionality of a data set by extracting a number of underlying factors, dimensions, clusters, etc., that can account for the variability in the (multidimensional) data set. For example, in poorly designed questionnaires, all responses provided by the participants on a large number of variables (scales, questions, or dimensions) could be explained by a very limited number of "trivial" or artifactual factors. For example, two such underlying factors could be: (1) the respondent's attitude towards the study (positive or negative) and (2) the "social desirability" factor (a response bias representing a tendency to respond in a socially desirable manner).

**Data Reduction by unbiased decreasing of the sample size (exploratory graphics).** This type of *Data Reduction* is applied in exploratory graphical data analysis of extremely large data sets. The size of the data set can obscure an existing pattern (especially in large line graphs or scatterplots) due to the density of markers or lines. Then, it can be useful to plot only a representative subset of the data (so that the pattern is not hidden by the number of point markers) to reveal the otherwise obscured but still reliable pattern. For an animated illustration, see the Data Reduction section of Selected Topics in Graphical Analytic Techniques.

Data Rotation (in 3D space). Changing the viewpoint for 3D scatterplots (e.g., simple, spectral, or space plots) may prove to be an effective exploratory technique since it can reveal patterns that are easily obscured unless you look at the "cloud" of data points from an appropriate angle (see the animation below).

Rotating or spinning a 3D graph will allow you to find the most informative location of the "viewpoint" for the graph. For more information see the section on Data Rotation (in 3D space) in Graphical Techniques.

Data Warehousing. StatSoft defines *data warehousing* as a process of organizing the storage of large, multivariate data sets in a way that facilitates the retrieval of information for analytic purposes.

For more information, see *Data Warehousing*.

Decision Trees. Decision trees are a class of predictive data mining tools which predict either a categorical or continuous response variable using a tree-like model. Several tools fall into the category of decision tree including Classification and Regression Trees (C&RT), Chi Square Automatic Interaction Detector (CHAID), Random Forests and Boosted Trees.

For more information, see Decision Trees.

Degrees of Freedom. Used in slightly different senses throughout the study of statistics, *Degrees of Freedom* were first introduced by Fisher based on the idea of degrees of freedom in a dynamical system (e.g., the number of independent co-ordinate values which are necessary to determine it). The degrees of freedom of a set of observations are the number of values which could be assigned arbitrarily within the specification of the system. For example, in a sample of size *n* grouped into *k* intervals, there are *k-1* degrees of freedom, because *k-1* frequencies are specified while the other one is specified by the total size *n*. Thus in a p by q contingency table with fixed marginal totals, there are (p-1)(q-1) degrees of freedom. In some circumstances the term *degrees of freedom* is used to denote the number of independent comparisons which can be made between the members of a sample.

Deleted Residual. The *deleted residual* is the residual value for the respective case, had it not been included in the regression analysis, that is, if one would exclude this case from all computations. If the *deleted residual* differs greatly from the respective standardized residual value, then this case is possibly an outlier because its exclusion changed the regression equation. See also, standard residual value, Mahalanobis distance, and Cook’s distance.

Denominator Synthesis. A method developed by Satterthwaite (1946) which finds the linear combinations of sources of random variation that serve as appropriate error terms for testing the significance of the respective effect of interest in mixed-model ANOVA/ANCOVA designs.

For descriptions of *denominator synthesis, *see *Variance Components and Mixed-Model ANOVA/ANCOVA* and *General Linear Models*.

Deployment. The concept of deployment in predictive data mining refers to the application of a model for prediction or classification to new data. After a satisfactory model of set of models have been identified (trained) for a particular application, one usually wants to deploy those models so that predictions or predicted classifications can quickly be obtained for new data. For example, a credit card company may want to deploy a trained model or set of models (e.g., neural networks, meta-learner) to quickly identify transactions which have a high probability of being fraudulent.

Derivative-Free Function Minimization Algorithms. Nonlinear Estimation offers several general function minimization algorithms that follow different search strategies which do not depend on the second-order derivatives. These strategies are sometimes very effective for minimizing loss functions with local minima.

Design Matrix. In *general linear models *and *generalized linear models*, the *design matrix* is the matrix ** X **for the predictor variables which is used in solving the normal equations.

*X**is a matrix, with 1 row for each case and 1 column for each coded predictor variable in the design, whose values identify the levels for each case on each coded predictor. See also general linear model, generalized linear model.*

Desirability Profiles. The relationship between predicted responses on one or more dependent variables and the desirability of responses is called the desirability function. Profiling the desirability of responses involves, first, specifying the desirability function for each dependent variable, by assigning predicted values a score ranging from 0 (very undesirable) to 1 (very desirable). The individual desirability scores for the predicted values for each dependent variable are then combined by computing their *geometric mean*. *Desirability profiles* consist of a series of graphs, one for each independent variable, of overall desirability scores at different levels of one independent variable, holding the levels of the other independent variables constant at specified values. Inspecting the *desirability profiles* can show which levels of the predictor variables produce the most desirable predicted responses on the dependent variables. For a detailed description of response/desirability profiling, see Profiling Predicted Responses and Response Desirability.

Detrended Probability Plots. This type of graph is used to evaluate the normality of the distribution of a variable, that is, whether and to what extent the distribution of the variable follows the normal distribution. The selected variable will be plotted in a scatterplot against the values "expected from the normal distribution." This plot is constructed in the same way as the standard normal probability plot, except that before the plot is generated, the linear trend is removed. This often "spreads out" the plot, thereby allowing the user to detect patterns of deviations more easily.

Deviance. To evaluate the goodness of fit of a generalized linear model, a common statistic that is computed is the so-called *Deviance *statistic. It is defined as:

Deviance = -2 * (Lm - Ls)

where *Lm* denotes the maximized log-likelihood value for the model of interest, and *Ls* is the log-likelihood for the saturated model, i.e., the most complex model given the current distribution and link function . For computational details, see Agresti (1996). See also the description of *Generalized Linear Models*.

Deviance residuals. After fitting a generalized linear model to the data, to check the adequacy of the respective model, one usually computes various residual statistics. The *deviance residual *is computed as:

r_{D} = sign(y-µ)sqrt(d_{i})

Where Sd_{i} = D, and D is the overall deviance measure of discrepancy of a generalized linear model (see McCullagh and Nelder, 1989, for details). Thus, the deviance statistic for an observation reflects its contribution to the overall goodness of fit (deviance) of the model. See also, *Generalized Linear Models*.

Deviation. In radial units, a figure multiplied by the radial exemplar's squared distance from the input pattern to generate the unit's activation level, before submission to the activation function. See neural networks.

Deviation Assignment Algorithms (in Neural Networks). These algorithms assign deviations to the radial units in certain network types. The deviation is multiplied by the distance between the unit's exemplar vector and the input vector, to determine the unit's output. In essence, the deviation gives the size of the cluster represented by a radial unit. Deviation assignment algorithms are used after radial centers have been set; see Radial Sampling and K Means.

Explicit Deviation Assignment. The deviation is set to an explicit figure provided by the user.

**Notes.** The deviation assigned by this technique is not the standard deviation of the Gaussians; it is the value stored in the unit threshold, which is multiplied by the distance of the weight vector from the input vector. It is related to the standard deviation by:

Isotropic Deviation Assignment. This algorithm uses the isotropic deviation heuristic (Haykin, 1994) to assign the deviations to radial units. This heuristic attempts to determine a reasonable deviation (the same for all units), based upon the number of centers, and how spread out they are.

This isotropic deviation heuristic sets radial deviations to:

where *d* is the distance between the two most distant centers, and k is the number of centers.

*k*-Nearest Neighbor Deviation. The *k*-nearest neighbor deviation assignment algorithm (Bishop, 1995) assigns deviations to radial units by using the RMS (Root Mean Squared) distance from the *k* units closest to (but not coincident with) each unit as the standard deviation (assuming the unit models a Gaussian). Each unit hence has its own independently calculated deviation, based upon the density of points close to itself. If less than *k* non-coincident neighbors are available, the algorithm uses the neighbors that are available.

Deviation Plots 3D. Data (representing the *X*, *Y*, and *Z* coordinates of each point) in this type of graph are represented in 3D space as "deviations" from a specified base-level of the *Z*-axis.

Deviation plots are similar to space plots. As compared to space plots, however in deviation plots the "deviations plane" is "invisible" and not marked by the location of the X-Y axes (those axes are always fixed in the standard bottom position). *Deviation plots* may help explore the nature of 3D data sets by displaying them in the form of deviations from arbitrary (horizontal) levels. Such "cutting" methods can help identify interactive relations between variables. See also, Data Rotation (in 3D space) in Graphical Techniques.

DFFITS. Several measures have been given for testing for leverage and influence of a specific case in regression (including studentized residuals, studentized deleted residuals, DFFITS, and standardized DFFITS). Belsley et al. (1980) have suggested *DFFITS*, a measure which gives greater weight to outlying observations than Cook's distance. The formula for *DFFITS* is

DFFIT_{i} = _{i}e_{i}/(1-_{i}) where

e_{i} is the error for the ith case

h_{i} is the leverage for the ith case

and _{i} = 1/N + h_{i}. For more information see Hocking (1996) and Ryan (1997).

DIEHARD Suite of Tests and Random Number Generation. Many areas of statistical analysis, research, and simulation rely on the quality of random number generators. Most programs for statistical data analysis contain a function for generating uniform random numbers. A recent review of statistical packages (McCullough, 1998, 1999) that appeared in *The American Statistician* tested the random number generators of several programs using the so-called *DIEHARD suite of tests* (Marsaglia, 1998). DIEHARD applies various methods of assembling and combining uniform random numbers, and then performs statistical tests that are expected to be nonsignificant; this suite of tests has become a standard method of evaluating the quality of uniform random number generator routines.

Differencing (in Time Series). In this Time Series transformation, the series will be transformed as: X=X-X(lag). After differencing, the resulting series will be of length *N-lag* (where *N* is the length of the original series).

Dimensionality Reduction. Data Reduction by decreasing the dimensionality (exploratory multivariate statistics). This interpretation of the term Data Reduction pertains to analytic methods (typically multivariate exploratory techniques such as Factor Analysis, Multidimensional Scaling, Cluster Analysis, Canonical Correlation, or Neural Networks) that involve reducing the dimensionality of a data set by extracting a number of underlying factors, dimensions, clusters, etc., that can account for the variability in the (multidimensional) data set. For more information, see Data Reduction.

Discrepancy Function. A numerical value that expresses how badly a structural model reproduces the observed data. The larger the value of the discrepancy function, the worse (in some sense) the fit of model to data. In general, the parameter estimates for a given model are selected to make a discrepancy function as small as possible.

The discrepancy functions employed in structural modeling all satisfy the following basic requirements:

- They are non-negative, i.e., always greater than or equal to zero.
- They are zero only if fit is perfect, i.e., if the model and parameter estimates perfectly reproduce the observed data.
- The discrepancy function is a continuous function of the elements of S, the sample covariance matrix, and (), the "reproduced" estimate of S obtained by using the parameter estimates and structural model.

Discriminant Function Analysis. *Discriminant function analysis* is used to determine which variables discriminate between two or more naturally occurring groups (it is used as either a hypothesis testing or exploratory method). For example, an educational researcher may want to investigate which variables discriminate between high school graduates who decide (1) to go to college, (2) to attend a trade or professional school, or (3) to seek no further training or education. For that purpose the researcher could collect data on numerous variables prior to students' graduation. After graduation, most students will naturally fall into one of the three categories. *Discriminant Analysis* could then be used to determine which variable(s) are the best predictors of students' subsequent educational choice (e.g., IQ, GPA, SAT).

For more information, see Discriminant Function Analysis; see also, Classification Trees.

Distributed File System. Big data (multiple terabytes, petabytes) can be stored and organized in distributed file systems. While there are multiple implementations and implementation details, in the most general terms, information is stored on one of multiple (sometimes thousands of) hard drives and standard off-the-shelf computers; an index or map keeps track of where (on which

computer/drive) a specific piece of information is stored. Actually, for failover redundancy and robustness, each piece of information is usually stored multiple times, e.g., as triplets.

So, for example, suppose you collected individual transactions in a large retail chain store. Details of each transaction would be stored in triplets on different servers and hard drives, with a master table or map keeping track of where exactly the respective transaction details can be retrieved. By using off-the-shelf standard hardware and open-source software for managing this distributed file system (such as Hadoop), reliable data repositories on the petabyte scale can relatively easily be achieved, and such storage systems are quickly becoming commonplace.

Document Frequency. The *document frequency* is a useful statistic computed for individual words or terms in the context of text mining. It denotes the number of documents in a collection of documents in which the respective word or term was found. For more information, see Manning and Schütze (2002).

Double-Y Histograms. The Double-Y histogram can be considered to be a combination of two separately scaled multiple histograms. Two different series of variables can be selected. A frequency distribution for each of the selected variables will be plotted but the frequencies of the variables entered into the first list (called *Left-Y variables*) will be plotted against the *left-Y* axis, whereas the frequencies of the variables entered into the second list (called *Right-Y variables*) will be plotted against the *right-Y* axis. The names of all variables from the two lists will be included in the legend followed by a letter *L* or *R*, denoting the *Left-Y* and *Right-Y* axis, respectively.

This graph is useful to compare distributions of variables with different frequencies.

Drill-Down Analysis. The concept of drill-down analysis applies to the area of data mining, to denote the interactive exploration of data, in particular of large databases. The process of drill-down analyses begins by considering some simple break-downs of the data by a few variables of interest (e.g., Gender, geographic region, etc.). Various statistics, tables, histograms, and other graphical summaries can be computed for each group. Next one may want to "drill-down" to expose and further analyze the data "underneath" one of the categorizations, for example, one might want to further review the data for males from the mid-west. Again, various statistical and graphical summaries can be computed for those cases only, which might suggest further break-downs by other variables (e.g., income, age, etc.). At the lowest ("bottom") level are the raw data: For example, you may want to review the addresses of male customers from one region, for a certain income group, etc., and to offer to those customers some particular services of particular utility to that group.

Duncan's Test. This post hoc test (or multiple comparison test) can be used to determine the significant differences between group means in an analysis of variance setting. *Duncan's* test, like the Newman-Keuls test, is based on the range statistic (for a detailed discussion of different post hoc tests, see Winer, Michels, & Brown (1991). For more details, see General Linear Models. See also, Post Hoc Comparisons. For a discussion of statistical significance, see Elementary Concepts.

Dunnett's test. This post hoc test (or multiple comparison test) can be used to determine the significant differences between a single control group mean and the remaining treatment group means in an analysis of variance setting. Dunnett's test is considered to be one of the least conservative post hoc tests (for a detailed discussion of different post hoc tests, see Winer, Michels, & Brown (1991). For more details, see General Linear Models. See also, Post Hoc Comparisons. For a discussion of statistical significance, see Elementary Concepts.

DV. *DV* stands for Dependent Variable. See also Dependent vs. Independent Variables.