Compute the critical value for the F-distribution, given the probability level, and the numerator and denominator degrees of freedom. Knowing the F-value for a particular probability level is often very useful in analytics.
Compute the 90%, 95%, and 99% confidence intervals for Cohen's f-square effect size for a multiple regression study, given the f-square value, the number of predictor variables, and the total sample size. Knowing the confidence interval for an f-square effect size can be very useful for comparing different models in analytics studies that rely on multiple regression.
Compute an adjusted R-square (or population R-square) value, given an observed (sample) R-square value, the number of predictor variables, and the total sample size. Knowing adjusted R-square values can be very useful in analytics for comparing models that use different numbers of predictor variables.
Compute Cohen's f-square effect size for a multiple regression study, given the study's R-square value. Effect sizes are often useful in analytics for quantifying the substantive value of the statistical model.
Compute the F-value and associated p-value for a multiple regression study, given the model R-square, the number of predictor variables, and the total sample size. F and p-values can be very useful ways of assessing and comparing different regression models when performing analytics.
Compute the observed power for your multiple regression study, given the observed p-value, the number of predictor variables, the observed R-square, and the sample size. When a regression model is not significant in an analytics study, it may be useful to know whether the model had sufficient power to detect an effect.
Compute the 90%, 95%, and 99% confidence intervals for an R-square value, given the R-square value, the number of predictor variables, and the total sample size. Knowing the confidence interval for an R-square value can be very useful in analytics when considering the true degree of usefulness that a regression model might have in the overall population.
Compute the minimum required sample size for your multiple regression study, given your desired p-value, the number of predictor variables in your model, the expected effect size, and your desired statistical power level. Knowing if your sample is large enough to detect an expected or hypothesized effect is critical to using multiple regression correctly in analytics.
Compute the Type 2 error rate (false negative rate) for your multiple regression study, given the observed p-value, the number of predictor variables, the observed R2, and the study's sample size. Knowing the chances of observing a false negative when using multiple regression is often very important in analytics.
Compute the 99%, 95%, and 90% confidence intervals for a predicted value of a regression equation, given the standard error of the estimate, the number of predictor variables, the total sample size, and a predicted value of the dependent variable. Knowing the confidence interval for a predicted regression value can be very useful for assessing the true range of outcomes that might occur in light of a given set of input values in analytics studies that rely on multiple regression.
Compute an R-square value for a multiple regression model, given the value of Cohen's f-square effect size for the model. Knowing the R-square value for a regression model is often very useful for assessing and comparing different regression models in analytics studies.
Compute 90%, 95%, and 99% confidence intervals for a regression coefficient, given the regression coefficient value, the standard error of the regression coefficient, the number of predictor variables, and the total sample size. Knowing the confidence interval for a regression coefficient can be very useful in analytics when considering the true range of values that a predictor variable might have in the overall population.
Compute the 90%, 95%, and 99% confidence intervals for a regression intercept (or regression constant), given the regression intercept value, the standard error of the regression intercept, the number of predictor variables, and the total sample size. Knowing the confidence interval for a regression intercept can be very useful in analytics when considering the true range of values that the intercept might have in the overall population.
Compute the extent to which the slopes of two lines are significantly different from one another, given each line's slope, standard error, and sample size. The calculator computes the t-value for the significance test, the degrees of freedom, and the p-value. Knowing whether the slopes of two lines are significantly different from one another can be very useful in analytics studies that compare multiple groups.