Presentation is loading. Please wait.

Presentation is loading. Please wait.

CPT tools Very much a work in progress. Manipulating CPTs numericPartSplits a mixed data frame into a numeric matrix and a factor part. rescaleTableRescales.

Similar presentations


Presentation on theme: "CPT tools Very much a work in progress. Manipulating CPTs numericPartSplits a mixed data frame into a numeric matrix and a factor part. rescaleTableRescales."— Presentation transcript:

1 CPT tools Very much a work in progress

2 Manipulating CPTs numericPartSplits a mixed data frame into a numeric matrix and a factor part. rescaleTableRescales the numeric part of the table normalizeTable scaleMatrixScales a matrix to have a unit diagonal scaleTableScales a table according to the Sum and Scale column. getTableParentsGets meta data about a conditional probability table. getTableStatesGets meta data about a conditional probability table.

3 Combination Rules/Structure Functions CompensatoryDiBello-Samejima combination function Conjunctive Disjunctive OffsetConjunctiveConjunctive combination function with one difficulty per parent. OffsetDisjunctive eThetaFrameConstructs a data frame showing the effective thetas for each parent combination. effectiveThetasAssigns effective theta levels for categorical variable

4 DiBello-XX Models calcDSTable Creates the probability table for DiBello-Samejima distribution calcDSFrame calcDNTableCreates the probability table for DiBello-Normal distribution calcDNFrame calcDDTableCalculates DiBello-Dirichlet model probability and parameter tables calcDDFrame calcDSllikeCalculates the log-likelihood for data from a DiBello- Samejima (Normal) distribution

5 Discrete Partial Credit Model calcDPCTableCreates the probability table for the discrete partial credit model calcDPCFrame gradedResponseA link function based on Samejima's graded response partialCreditA link function based on the generalized partial credit model mapDPCFinds an MAP estimate for a discrete partial credit CPT

6 Noisy-logic models calcNoisyAndTableCalculate the conditional probability table for a Noisy-And or Noisy-Min distribution calcNoisyAndFrame calcNoisyOrTableCalculate the conditional probability table for a Noisy-Or or Noisy-Max distribution calcNoisyOrFrame

7 Model Construction Utilities buildFactorTabBuilds probability tables from Scored Bayes net output. build2FactorTab buildMarginTab marginTab buildParentListBuilds a list of parents of nodes in a graph dataTableConstructs a table of counts from a set of discrete observations. mcSearchOrders variables using Maximum Cardinality search structMatrixFinds graphical structure from a covariance matrix

8 Normal Model Utilities areaProbsTranslates between normal and categorical probabilities pvecToCutpoints pvecToMidpoints buildRegressionTablesBuilds conditional probability tables from regressions buildRegressionsCreates a series of regressions from a covariance matrix

9 Output Plots colorspreadProduces an ordered palate of colours with the same hue. stackedBarsProduces a stacked, staggered barplot stackedBarplot compareBarsProduces comparison stacked bar charts for two sets of groups compareBars2 parseProbVecParses Probability Vector Strings

10 Diagnostic Plots & Tests OCP2Observable Characteristic Plot OCP betaciCredibility intervals for a proportion based on beta distribution proflevelciProduce cumulative sum credibility intervals ciTest localDepTest

11 Weight-of-Evidence mutualInformationCalculates Mutual Information for a two-way table. readHistoryReads a file of histories of marginal distributions. woeBalWeight of Evidence Balance Sheet woeHistCreates weights of evidence from a history matrix.

12 Data Sets MathGradesGrades on 5 mathematics tests from Mardia, Kent and Bibby


Download ppt "CPT tools Very much a work in progress. Manipulating CPTs numericPartSplits a mixed data frame into a numeric matrix and a factor part. rescaleTableRescales."

Similar presentations


Ads by Google