多元回归

  • Cell Phone Service

    Multiple Regression - Two Predictors

    Determine whether wind speed and barometric pressure are related to phone call performance (percentage of dropped or failed calls).

    JMP features demonstrated:

    Analyze > Distribution, dynamic plot linking, Analyze > Fit Y by X (Bivariate), Analyze > Fit Model (Standard Least Squares), Analyze > Multivariate Methods > Multivariate, Surface Profiler (from Fit Model)

    Statistical/Graphical Tools Used:

    Histograms, summary statistics, simple linear regression, multiple regression, scatterplots, and three-dimensional scatterplot.

  • Housing Prices

    Multiple Regresion - Multicollinearity and Model Building

    After determining which factors relate to the selling prices of homes located in and around a ski resort, develop a model to predict housing prices.

    JMP features demonstrated:

    Analyze > Fit Model, Analyze > Multivariate Methods > Multivariate.

    Statistical/Graphical Tools Used:

    Scatterplot matrix, pairwise and partial correlations, multiple regression, VIFs, stepwise regression, model diagnostics.

  • Lost Sales

    Logistic Regression

    Determine whether certain conditions make it more likely that a customer order will be won or lost.

    JMP features demonstrated:

    Analyze > Distribution, Analyze > Fit Y by X (Contingency), Analyze > Fit Y by X (Logistic), Analyze > Fit Model (Nominal Logistic), and Prediction Profiler (from Fit Model)

    Statistical/Graphical Tools Used:

    Bar charts and frequency distributions, mosaic plots, contingency tables (cross tabs), chi-squared tests, logistic regression, predicted values and confusion matrix.

Back to Top