Statistical Analysis - Aster Analytics

Teradata AsterĀ® Analytics Foundation User GuideUpdate 2

Product
Aster Analytics
Release Number
7.00.02
Published
September 2017
Language
English (United States)
Last Update
2018-04-17
dita:mapPath
uce1497542673292.ditamap
dita:ditavalPath
AA-notempfilter_pdf_output.ditaval
dita:id
B700-1022
lifecycle
previous
Product Category
Software
Statistical Analysis Functions
Function Description
Approximate Distinct Count Computes the approximate global distinct count of the values in one or more columns, scanning the table only once. Counts all children for a specified parent.
Approximate Percentile Computes approximate percentiles for one or more columns, with specified accuracy.
ConfusionMatrix Shows how often a classification algorithm correctly classifies items.
Correlation Computes the global correlation between any pair of table columns.
CoxPH Estimates coefficients of a Cox proportional hazards model by learning a set of explanatory variables. Generates coefficient and linear prediction tables.
CoxPredict Takes the coefficient table generated by the CoxPH function and outputs the hazard ratios between predict features and either their corresponding reference features or their unit differences.
CoxSurvFit Takes the coefficient and linear prediction tables generated by the CoxPH function and outputs a table of survival probabilities.
CrossValidation Validates a model by assessing how the results of a statistical analysis generalize to an independent data set.
Distribution Matching Uses hypothesis testing to find the best matching distribution for data.
FMeasure Calculates the accuracy of a test.
Generalized Linear Model Functions Perform linear regression analysis for distribution functions using a user-specified distribution family and link function. The generalized linear model (GLM) functions are GLM, GLMPredict, GLM2, and GLM2Predict.
Hidden Markov Model Functions Describes the evolution of observable events that depend on factors that are not directly observable. The Hidden Markov Model functions are HMMUnsupervisedLearner, HMMSupervisedLearner, HMMEvaluator, and HMMDecoder.
Histogram (Hist) Calculates the frequency distribution of a data set using sophisticated binning techniques that can automatically calculate the bin width and number of bins. The function maps each input row to one bin and returns the frequency (row count) and proportion (percentage of rows) of each bin.
KNN Uses the kNN algorithm to classify new objects based on their proximity to already-classified objects.
LARS Functions Selects the most important variables one by one and fit the coefficients dynamically. The LARS functions are LARS and LARSPredict.
Linear Regression Outputs the coefficients of the linear regression model represented by the input matrices.
LinRegPredict Takes a model built by Linear Regression and a test data set whose input attributes are the same as those in the model, and predicts the response variable for each observation in the test data set.
LRTEST Performs the likelihood ratio test for two GLM models.
Moving Average Functions Compute average values in a series. The moving average functions are CMAVG, EMAVG, SMAVG, and WMAVG.
Percentile Finds percentiles on a per group basis.
Principal Component Analysis (PCA) Common unsupervised learning technique that is useful for both exploratory data analysis and dimensionality reduction, often used as the core procedure for factor analysis. Implemented by the functions PCA_Map and PCA_Reduce. If the version of PCA_Reduce is AA 6.21 or later, you can input the PCA output to the function PCAPlot.
RandomSample Takes a data set and uses a specified sampling method to output one or more random samples, each with a specified size.
Receiver Operating Characteristic (ROC) Takes a set of prediction-actual pairs for a binary classifier and calculates the TPR, FPR, AUC, and Gini coefficient for a range of thresholds.
Sample Draws rows randomly from input, using either of two sampling schemes.
Shapley Value Functions Computes the Shapley value, typically from nPath function output. The Shapley value is intended to reflect the importance of each player to the coalition in a cooperative game (a game between coalitions of players, rather than between individual players). The Shapley value functions are GenerateCombination, SortCombination, and AddOnePlayer.
Support Vector Machine (SVM) Functions Uses a popular classification algorithm to build a predictive model according to a training set, give a prediction for each sample in the test set, and display the readable information of the model. The SparseSVM Functions are SparseSVMTrainer, SparseSVMPredictor, and SVMModelPrinter. The DenseSVM Functions are DenseSVMTrainer, DenseSVMPredictor and DenseSVMModelPrinter.
VectorDistance Measures the distance between sparse vectors (for example, TF-IDF vectors) in a pairwise manner.
VWAP Computes the volume-weighted average price of a traded item (usually an equity share) over a specified time interval.