LoadingLoading...

Regression Analysis in NCSS

Regression analysis refers to a group of techniques for studying the relationships among two or more variables. NCSS makes it easy to run either a simple linear regression analysis or a complex multiple regression analysis. You can perform a regression analysis with modern graphical and numeric residual analysis. Major options include multiple regression, stepwise regression, correlation matrix, residual analysis, robust regression, all-possible regressions, response surface regression, Poisson regression, Cox regression, ridge regression, Weibull regression, and logistic regression.

t_productimg_293

Features

> All-possible regressions
> Alpha level is flexible
> Beta coefficients
> Binary variables are automatic
> Biweight regression
> Bootstrap Confidence Intervals
> Categorical variables
> Coefficient of variation
> Condition number
> Confidence limits
> Cook’s D
> Correlations
> CovRatio
> Cp Statistic
> DFBetas
> Dffits
> Durbin Watson statistic
> Eigenvalues
> Eigenvectors
> F-ratios
> Hat diagonal
> Hierarchal model search
> Huber’s robust regression
> Interactions automatic
> LAV regression
> Logistic regression
> Multinomial logistic regression
> Multiple regression
> Normal probability plot
> Normality tests of residuals
> R-squared
> Regression coefficient
> Residual analysis and plots
> Response surface analysis
> Ridge regression
> Robust regression
> Rstudent
> Serial correlation plot
> Simple linear regression
> Spearman’s rank correlation
> Standard error of beta
> Standardized coefficient
> Stepwise regression
> Sum of squares
> T-test for beta=0
> Through the origin
> Tolerance level
> Variable selection
> Variance inflation factor
> Weighted regression
> Partial correlation
> Pearson’s correlation
> Poisson regression
> Power calculations
> Predicted values
> Prediction limits
> Predicts new observations
> Press statistic
> P.C. regression
> Proportional hazards
> R-bar squared
123