Ridge regression techniques have been extensively used to solve the multicollinearity problem for both linear and non-linear regression models since its inception. This paper studied different ridge regression t-type tests of the individual coefficients of a linear regression model. A simulation study has been conducted to evaluate the performance of the proposed tests with respect to their sizes and powers under different settings of the linear regression model. Our simulation results demonstrated that most of the proposed tests have sizes close to the 5% nominal level and all tests except tAKS, tkM2 and tkM9 have considerable gain in powers over the ordinary OLS t-type test. It is also observed that some of the proposed test statistics are performing better than the HK and HKB tests which are proposed some authors.
Golam Kibria, B. M. and Banik, Shipra
A Simulation Study on the Size and Power Properties of Some Ridge Regression Tests,
Applications and Applied Mathematics: An International Journal (AAM), Vol. 14,
2, Article 7.
Available at: https://digitalcommons.pvamu.edu/aam/vol14/iss2/7