Relative Importance of Openness Index and Crude Oil Price in Explaining Growth of India's GDP During 1970-71 to 2019-20
Keywords:
True Relative Importance, Ortho-partial Correlation, GDP. JEL Classification- C18, F43Abstract
This paper seeks to evaluate the relative importance of the Openness Index (OPI) and Crude Oil Price (COP) to explain the growth of the GDP of India in the period from 1970-71 to 2019-20. For this purpose, this study uses a very popular and important method of evaluating the relative importance of predictors in the context of linear multiple regression analysis, viz., Budescu’s Average Dominance (BAD) method (Budescu [1]. This study observes that two aspects of multicollinearity, viz., enhancement-synergism (ES) and change-in-sign (CS), not properly addressed in the existing literature, though observed in the majority of illustrations, are also not well taken by the above method. As a result, mainly for the second aspect, BAD methods lead to an improper evaluation of their relative importance. Under these circumstances, this study proposes a new method for evaluating True Relative Importance (TRI) of the predictors that is also able to assign their role. Using this new method this paper makes the concluding observation that openness index is more important and significant factor stimulating growth than crude oil price.
Downloads
References
Budescu, D.V. (1993), Dominance analysis: A new approach to the problem of relative importance of predictors in multiple regression. Psychological Bulletin, 114(3), 542– 551 https://doi.org/10.1037/0033-2909.114.3.542
Chao YC, Zhao Y, Kupper LL, & Nylander-French LA. (2008), Quantifying the Relative Importance of Predictors in Multiple Linear Regression Analyses for Public Health Studies', Journal of Occupational and Environmental Hygiene, 5:8, 519-529, https://doi.org/10.1080/15459620802225481
Hamilton, D.C. (1987), Sometimes R2 > ryx12 + ryx22 - Correlated Variables are not Always Redundant, The American Statistician, 41(2), 129-132. https://doi.org/10.1080/00031305.1987.10475460
Mondal, D., (2008), On the test of significance of linear multiple regression coefficients, Communications in Statistics: Simulation and Computation, Vol. 37(4), https://doi.org/10.1080/03610910701753853