ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

An Econometric Analysis of World GDP Share of India during 1960-2015

Journal: SocioEconomic Challenges (SEC) (Vol.2, No. 1)

Publication Date:

Authors : ;

Page : 40-55

Keywords : international GDP share; exponential growth; structural break; non-stationary; HP filter; peaks; trough;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

In this paper author attempted to analyse India's international GDP share during 1960-2015 with the help of econometric models taking the data from the World Bank. Semilog linear trend model and ex-ponential trend model were used to find the trend of growth. Variance ratio test was used to show ran-dom walk. AR(1) model was used to show stationary, convergence and oscillations. ARIMA (1,1,1) model was tested for the stationary of the series. Forecast for 2035 of the AR(1) and ARIMA(1,1,1) models verified stationary long term patterns. Bai-Perron (2003) model explained to show structural breaks and the study of Bartoletto, Chiarini, Marzano & Piselli (2015) was followed to compute peaks, troughs, durations of cycles, amplitudes and slopes of both the short and medium cycles during 1960-2015. Hodrick-Prescott Filter (1997) model minimized the cycles for smoothness of trend of GDP share.The paper concludes that international GDP share of India has decreased at the rate of 0.459% per year during 1960-2015 and declined exponentially at the rate 0.259% per year significantly. The growth rate of the GDP share is downward sloping significantly till 2030. It follows random walk without drift. Its AR(1) is stable, convergence and stationary. Forecast for 2035 of AR(1) is also con-verging. ARIMA(1,1,1) is stable and non-stationary and suffers from AC and PAC problems. Its fore-cast model for 2035 is tending towards stationary insignificantly. GARCH (1,1) showed excessive volatility. It has two downward structural breaks in 1968 and 1988 and one upward break in 2006 which are significant. The paper verified short and medium cycles to calculate peaks and troughs, du-ration of downturn and upturn, amplitude and slope of the cycles respectively. HP filter model makes the cycle more smooth with only one trough assuming lamda comprises 1600 but symmetric and asymmetric filter showed two peaks and two troughs. The frequency response function clarified its peaks and amplitude of cycle clearly.

Last modified: 2018-04-12 20:59:46