Multinomial Logistic Regression: Complete ProblemsOutliers75/25% Cross-validation StrategySlide 4Problem 1Dissecting problem 1 - 1Dissecting problem 1 - 2Dissecting problem 1 - 3Dissecting problem 1 - 4Dissecting problem 1 - 5LEVEL OF MEASUREMENT - 1LEVEL OF MEASUREMENT - 2Request multinomial logistic regression for baseline modelSelecting the dependent variableSelecting metric independent variablesSpecifying statistics to include in the outputRequesting the classification tableCompleting the multinomial logistic regression requestClassification accuracy for all casesOutliers for the comparison of groups 1 and 3Selecting groups 1 and 3Formula for selecting groups 1 and 3Completing the selection of groups 1 and 3Binary logistic regression comparing groups 1 and 3Dependent and independent variables for the comparison of groups 1 and 3Including standardized residuals in the comparison of groups 1 and 3Slide 27Locating the case ids for outliers for groups 1 and 3Replace the selection criteriaFormula for identifying outliersCompleting the selection of outliersLocating the outliers in the data editorThe outliers in the data editorOutliers for the comparison of groups 2 and 3Selecting groups 2 and 3Formula for selecting groups 2 and 3Completing the selection of groups 2 and 3Binary logistic regression comparing groups 2 and 3Slide 39Locating the case ids for outliers for groups 2 and 3Slide 41Slide 42Slide 43Slide 44Slide 45The caseid of the outliersExcluding the outliers from the multinomial logistic regressionChanging the condition for the selectionExcluding cases 20002045 and 20002413Completing the exclusion of the outlierMultinomial logistic regression excluding the outlierRunning the multinomial logistic regression without the outlierClassification accuracy after omitting outliersRestoring the outlier to the data setRestoring the outliers to the data setRe-running the multinomial logistic regression with all casesRequesting the multinomial logistic regression againSAMPLE SIZE: ratio of cases to variablesOVERALL RELATIONSHIP BETWEEN INDEPENDENT AND DEPENDENT VARIABLESNUMERICAL PROBLEMSRELATIONSHIP OF INDIVIDUAL INDEPENDENT VARIABLES TO DEPENDENT VARIABLE - 1RELATIONSHIP OF INDIVIDUAL INDEPENDENT VARIABLES TO DEPENDENT VARIABLE - 2RELATIONSHIP OF INDIVIDUAL INDEPENDENT VARIABLES TO DEPENDENT VARIABLE - 3RELATIONSHIP OF INDIVIDUAL INDEPENDENT VARIABLES TO DEPENDENT VARIABLE - 4RELATIONSHIP OF INDIVIDUAL INDEPENDENT VARIABLES TO DEPENDENT VARIABLE - 5CLASSIFICATION USING THE MULTINOMIAL LOGISTIC REGRESSION MODEL: BY CHANCE ACCURACY RATECLASSIFICATION USING THE MULTINOMIAL LOGISTIC REGRESSION MODEL: CLASSIFICATION ACCURACYValidation analysis: set the random number seedSet the random number seedValidation analysis: compute the split variableThe formula for the split variableSelecting the teaching sampleSlide 73Slide 74Slide 75Re-running the multinomial logistic regression with the teaching sampleSlide 77Comparing the teaching model to full model - 1Comparing the teaching model to full model - 2Comparing the teaching model to full model - 3Classification accuracy of the holdout sampleThe log of the odds for the first groupThe log of the odds for the second groupThe log of the odds for the third groupThe probabilities for each groupGroup classificationSelecting the holdout sampleSlide 88Slide 89Slide 90The classification accuracy tableSlide 92Slide 93Slide 94The classification accuracy tableAnswering the question in problem 1 - 1Answering the question in problem 1 - 2Answering the question in problem 1 - 3Steps in multinomial logistic regression: level of measurementSteps in multinomial logistic regression: detecting outliersSteps in multinomial logistic regression: picking model for interpretationSteps in multinomial logistic regression: sample sizeSteps in multinomial logistic regression: overall relationship and numerical problemsSteps in multinomial logistic regression: relationships between IV's and DVSteps in multinomial logistic regression: split-sample validationSteps in multinomial logistic regression: validation supports generalizabilitySteps in multinomial logistic regression: adding cautionsSW388R7Data Analysis & Computers IISlide 1Multinomial Logistic Regression:Complete ProblemsOutliersSplit-sample ValidationSample ProblemsSW388R7Data Analysis & Computers IISlide 2OutliersMultinomial logistic regression in SPSS does not compute any diagnostic statistics. In the absence of diagnostic statistics, SPSS recommends using the Logistic Regression procedure to calculate and examine diagnostic measures.A multinomial logistic regression for three groups compares group 1 to group 3 and group 2 to group 3. To test for outliers, we will run two binary logistic regressions, using case selection to compare group 1 to group 3 and group 2 to group 3.From both of these analyses we will identify a list of cases with standardized residuals greater than ± 3.29, and test the multinomial solution without these cases. If the accuracy rate of this model is less than 2% more accurate, we will interpret the model that includes all cases.SW388R7Data Analysis & Computers IISlide 375/25% Cross-validation StrategyIn this validation strategy, the cases are randomly divided into two subsets: a training sample containing 75% of the cases and a holdout sample containing the remaining 25% of the cases.The training sample is used to derive the multinomial logistic regression model. The holdout sample is classified using the coefficients for the training model. The classification accuracy for the holdout sample is used to estimate how well the model based on the training sample will perform for the population represented by the data set. While it is expected that the classification accuracy for the validation sample will be lower than the classification for the training sample, the difference (shrinkage) should be no larger than 2%.In addition to satisfying the classification accuracy, we will require that the significance of the overall relationship and the relationships with individual predictors for the training sample match the significance results for the model using the full data set.SW388R7Data Analysis & Computers IISlide 475/25% Cross-validation StrategySPSS does not classify cases that are not included in the training sample, so we will have to manually compute the classifications for the holdout sample if we want to use this strategy.We will run the analysis for