| Autor: |
Yousuf, Rizwan, Sharma, Manish, Bhat, Anil |
| Předmět: |
|
| Zdroj: |
International Journal of Agricultural & Statistical Sciences; 2021 Suppl, Vol. 17, p2065-2072, 8p, 6 Charts |
| Abstrakt: |
For agronomic and agricultural economic modelling and analysis, a good representation of response functions is vital. In this article, the perspective is considered by different regression methods. This is due to the fact that if least squares regression is used, observations (outliers) might lead to misleading conclusions. Robust regression approaches such as M estimation, MM estimation, S estimation, and LTS estimation are used to solve this problem since they are unaffected by outliers. The quadratic function (QF) and Square root function (SF) are used to study the production as a response variable with labour and capital as explanatory variables of the maize data. The use of robust regression rather than least square regression results in more appropriate shifts in coefficient estimates and their degree of significance, as well as higher levels of goodness of fit. The proposed robust method after handling the outliers having the same precision of M estimation. The QF used to study the maize production found to be best as compared to the SF on the basis of cross validation techniques. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
