Box-Cox Normality Plot | |
# observations x | 13 |
maximum correlation | 0.985353792435655 |
optimal lambda | -0.36 |
transformation formula | for all lambda <> 0 : T(Y) = (Y^lambda - 1) / lambda |
Obs. | Original | Transformed |
1 | 11.24438088 | 1.61536412565273 |
2 | 17.66485036 | 1.78982107371322 |
3 | 15.18527714 | 1.73453516258741 |
4 | 25.30595776 | 1.90974265351736 |
5 | 24.00168259 | 1.89304831223488 |
6 | 94.14444444 | 2.23685980924587 |
7 | 43.37222 | 2.06278557696792 |
8 | 58.35 | 2.13520430974151 |
9 | 50.15652 | 2.09923124363076 |
10 | 17.17381 | 1.7797433935114 |
11 | 29.275 | 1.95409760140131 |
12 | 18.62142857 | 1.80840053296032 |
13 | 50.77307692 | 2.10220919380743 |
Maximum Likelihood Estimation of Lambda |
> summary(mypT) bcPower Transformation to Normality Est Power Rounded Pwr Wald Lwr Bnd Wald Upr Bnd x -0.2896 0 -1.2577 0.6784 Likelihood ratio test that transformation parameter is equal to 0 (log transformation) LRT df pval LR test, lambda = (0) 0.3482057 1 0.55513 Likelihood ratio test that no transformation is needed LRT df pval LR test, lambda = (1) 7.041686 1 0.0079634 |