Studies have shown that quadratic equation gives improved correlation coefficents than linear equation. Still, Linear correlation is more popular amung the modeling community.
What are the advantages of linear equation in relation to predictions that polynomial would not have?
Dear Mohd,
I suggest you to see links and attached file on topic.
Comparative Qsar - Page 19 - Résultats Google Recherche de Livres https://books.google.dz/books?isbn=1560327162
QSPR/QSAR Studies of 2-Furylethylenes Using Bond-Level Quadratic ...
www.scielo.org.mx/scielo.php?script=sci_arttext&pid=S1870...
Best regards
I got the answer.
It is assumed that the drug properties are real number and that they are independent of each other; so that the effects of one variable is the same regardless of the values of the other variable. In such situation, linear equation outperforms polynomial. Moreover, linear regression is found to be more robust and successful in forming the model. Studies have also shown that polynomial equations are statistically poorly characterized and can be slow and are difficult to interpret their meaning in chemical terms.
References
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling