Research Article
A New Look at Worst Case Complexity: A Statistical Approach
Box 5
Regression analysis result for
versus
and
(Box-Cox transformation) on random data sequences (5 * 10
5 ≤
≤ 50 * 10
5).
| General Regression Analysis: versus , | | Box-Cox transformation of the response with specified lambda = 0.5 | | Regression Equation | | 0.5=2.39481 − 4.2737e − 006 n + 2.34136e − 009 | | Coefficients | | TermCoefSE CoefTP | | Constant2.394810.8332612.874020.024 | | n−0.000000.000001−3.955140.005 | | 0.000000.0000005.306210.001 | | Summary of Model | | S = 0.606531R-Sq = 95.26%R-Sq(adj) = 93.90% | | PRESS = 9.71668R-Sq(pred) = 82.10% | | Analysis of Variance | | SourceDFSeq SSAdj SSAdj MSFP | | Regression2 51.707251.707225.853670.27740.0000233 | | n141.34935.75485.754815.64320.0054957 | | 110.357910.357910.357928.15580.0011158 | | Error72.57522.57520.3679 | | Total954.2824 | | Fits and Diagnostics for Unusual Observations for Transformed Response | | Obs0.5FitSE FitResidualSt Resid | | 108.058417.203520.4627070.8548952.18002R | | Fits for Unusual Observations for Original Response | | ObsyFit | | 1064.93851.8907R | | R denotes an observation with a large standardized residual. |
|
