If R1 is an array with multiple columns then the same approach is used except that now the x i represent rows in R1. The bootstrap sample is then a column array containing the elements f( S 1), …, f( S iter). If R1 is a column array with elements S =, then bootstrapping works by creating iter data sets S 1, …, S iter where each S j is formed by taking n elements from S with replacement. JACKKNIFE(R1, expression, ref): returns a column array with the jackknife sample for the data in R1 based on the function f( arr) on a variable arr that takes array values expression is used to specify f and ref is an optional reference to arr. This release adds general jackknife support for any parameter using the Real Statistics lambda capability. The Real Statistics Resource Pack currently supports jackknifing to estimate the standard error and confidence intervals for Deming regression and the Gini Index.
See TLS Regression Confidence Interval for more information. This release adds a new worksheet array function TRegPARAM that returns the coefficient, standard error, and lower/upper ends of the 1- α confidence interval for the intercept and slope coefficients based on total linear regression. See Moderation Analysis for more information. This tool is now included as an option on the Reg tab of the main menu. This release adds a new data analysis tool that performs Moderation Analysis.
The following is an overview of the new features in Release 8.3.