Table 4. Summary of results of statistical analysis and model equations for the measured responses.
| Models | R2 a | R2 (adj)b | SEc | PRESSd | Remark | |
|---|---|---|---|---|---|---|
| Droplet size (nm; Y1) | ||||||
| Linear | 79.93 | 78.26 | 24.75 | 17996.4 | ||
| Quadratic | 86.53 | 83.33 | 21.67 | 15288 | ||
| Special cubic | 88.54 | 85.11 | 20.49 | 13960.5 | ||
| Full cubic | 98.13 | 97.3 | 8.72 | 3077.3 | Suggested | |
| TMS solubilization (mg/mL; Y2) | ||||||
| Linear | 55.08 | 51.34 | 0.233896 | 1.68881 | ||
| Quadratic | 76.73 | 71.19 | 0.179977 | 1.09605 | ||
| Special cubic | 77.7 | 71.01 | 0.180531 | 1.14689 | ||
| Full cubic | 89.39 | 84.68 | 0.131233 | 0.6975 | Suggested | |
| Percentage of drug released in 15 minutes (%; Y3) | ||||||
| Linear | 2.08 | 0 | 11.4817 | 4090.59 | ||
| Quadratic | 70.59 | 63.59 | 6.72686 | 1398.17 | ||
| Special cubic | 77.87 | 71.23 | 5.97921 | 1170.12 | ||
| Full cubic | 96.83 | 95.42 | 2.38522 | 230.415 | Suggested | |
aPercentage of response variable variation; the higher the value, the better the model fits the data.
bPercentage of response variable variation based on its relationship with one or more predictor variables
cStandard error of the regression, represents the standard distance between the data values and the fitted regression line.
dPrediction error sum of squares, the smaller the PRESS value, the better the model predictive ability.