In ANOVA, if the null hypothesis is true, the F-statistic follows an F-distribution. The F-distribution is a probability distribution that is used most commonly in Analysis of Variance.

Spearman's Rank Correlation is especially useful when the relationship between variables is monotonic, but not necessarily linear. A monotonic relationship is one where the variables tend to change together, but not necessarily at a constant rate.

Interval estimation is generally preferred over point estimation because it provides a range of possible values rather than a single value. This range of values gives a better understanding of the uncertainty around the estimate, hence, it provides more information than a single point estimate.

The Wilcoxon Signed Rank test is a non-parametric test that compares the medians of two paired groups.

The support of a random variable is the set of values in the range of the variable that have non-zero probability. For a continuous random variable, it's the set of values over which the probability density function is non-zero.

The 'R-squared' value, also known as the coefficient of determination, in a multiple linear regression model represents the proportion of variance in the dependent variable that can be predicted from the independent variables. It ranges from 0 to 1, where a higher value indicates a better fit of the model.

The difference between kurtosis and excess kurtosis comes down to a constant. Excess kurtosis is simply kurtosis minus 3. The "3" comes from the kurtosis of a normal distribution which is 3. Hence, excess kurtosis refers to kurtosis in relation to a normal distribution.

In an ANOVA test, the F-statistic is calculated as the ratio of the between-group variance (mean sum of squares between groups) to the within-group variance (mean sum of squares within groups). A larger F-statistic implies a greater degree of difference between the group means.

Autocorrelation, also known as serial correlation, is the correlation of a signal with a delayed copy of itself as a function of delay. In the context of a regression analysis, it refers to the condition when the error term (residuals) in a regression model is correlated with itself.

By definition, the probability of an impossible event is 0. This is because the measure of probability assigns 0 to events that cannot occur and 1 to events that are certain to occur.

The effect size is the magnitude of the difference between groups. Larger effect sizes are easier to detect and are associated with higher power in a t-test.

To adjust for multicollinearity in a multiple linear regression model, one of the common strategies is to remove one of the highly correlated independent variables or to combine the correlated variables.