Seminar: "Exchange rates in India: Current Account Monetarism in a nonlinear context" by Dr. Aditi B Chaubal
The Indian exchange rate system has evolved from a pegged system to the current managed float. This study examines the presence of a long-run equilibrium in Indian exchange rates while accounting for different nonlinearities. The study examines the monthly Indian nominal exchange rate (Rs/USD) over the period January 1993 to January 2014 using monetary models viz. (i) current account monetarism (CAM) model or flexible price monetary model (FPMM) wherein exchange rates are determined by levels of output, short-run interest rates and money supply; and (ii) capital account monetarism (KAM) model or sticky price monetary model (SPMM) wherein the assumption of perfectly flexible prices is relaxed. The nonlinear adjustment to disequilibria is modelled using a nonlinear error correction model (NLECM) (Dufrénot and Mignon (2002), Chaouachi et al. (2004)) in the monetary model framework. The nonlinear CAM model implemented includes nonlinear transformations (transition functions) of long-run dynamics in the vector error correction model (VECM) to account for different nonlinearities viz. multiple equilibria (cubic polynomial function), nonlinear mean reversion (exponential smooth transition autoregressive function (ESTAR)), smooth and gradual regime switches (ESTAR) which may be present in the exchange rates. The NLECM estimated using the ESTAR transition function outperforms other alternatives including the random walk model in explaining Indian exchange rates, based on various model performance measures such as Akaike Information Criterion (AIC), adjusted R-squared, and forecast performance measures such as root mean squared error (RMSE) and mean absolute error (MAE). This implies the existence of nonlinear mean reversion and smooth transition across different periods of overvaluation and undervaluation of the exchange rate. A similar KAM specification could not be implemented due to exchange rate disequilibrium being stationary. The study estimates an appropriate Autoregressive Distributed Lags (ARDL) KAM model which proved the presence of long-run relationship disequilibria of exchange rates and prices but does not outperform a benchmark random walk model.
Dr. Aditi B Chaubal's primary research interest is the empirical applications of nonlinear time series models in Economics. This includes predictive modeling of economic and financial data in various ways such as classifying it or studying its transitional phases, to help augmenting formulation of economic policy. This also involves theorizing the academic aspects of applying new research tools to existing datasets and understanding the implications of their results. He was worked as a research assistant in the field of agriculture economics during the course of my PhD, and with the Bill and Melinda Gates Research Foundation project at Indira Gandhi Institute of Development Research (IGIDR), during which he was published a joint paper and a co-authored book. This has also made him adept at handling large datasets such as the ones in the National Sample Survey (NSS). He was completed his PhD in Economics from IGIDR. Before this, He completed his Master of Science in Economics from IGIDR in 2009 and, Bachelor of Science in Mathematics with double honors in Mathematics and Physics from St Xavier’s College, Mumbai in 2007. He submitted his PhD thesis in 2015 and was awarded the degree in 2017. His thesis titled ‘Essays on Nonlinear Times Series and its applications’ under Prof. Dilip Nachane (former Director, IGIDR), examined and analyzed the dynamics of Indian inflation and exchange rates using nonlinear time series models. He classified inflation using these models into regimes based on their long-run trends and analyzed its determinants in each phase. For exchange rates, he examined the long-run trend between exchange rates and their monetary determinants and found smooth exponential transition in the short-run across periods of overvaluation and undervaluation. In his current role at the Australian Consulate General in Mumbai, he analyze and report on Indian economic, political, and sectoral developments from an Australian perspective. He was sponsored by the Australian government to Canberra as an Indian expert to present an assessment of the prospects of growth and development of the Indian economy at their Prime Minister’s Office and attend a trade policy course by the foreign affairs department. He also a part of the India Economic Strategy working group in the Consulate. He continues his active work and publish research in addition to his current job. He have published a joint paper (The Plutocratic Bias in the Indian CPI) in The International Labour Review. He presented a paper at the 3rd International Workshop on Financial Markets and Nonlinear Dynamics in Paris in June 2017. It is currently being revised for re submission in the journal Studies in Nonlinear Dynamics and Econometrics . He like teaching and research and as such he is committed to academics and have always aspired to be in an academic field. He enjoyed teaching a refresher statistics course for 3 years to entry-level Masters students, and EViews software for time series analysis for 2 years to Reserve Bank of India officers. He intend to continue carrying out predictive modeling of macroeconomic datasets to garner insights into forming scalable economic policies and strive to keep learning various applications of mathematical techniques in economics.