Necessity of Incorporating Break-in-trend in Aggregate Time Series Data: Evidence from Agricultural GDP in India

Vol 11 , Issue 2 , July - December 2024 | Pages: 27-38 | Research Paper

Article has been added to the cart.View Cart (0)

https://doi.org/10.17492/jpi.pragati.v11i2.1122402

Published Online: October 03, 2024

Author Details ( * ) denotes Corresponding author

1. Debasish Mondal, Professor, Economics, Vidyasagar University, Midnapore, West Bengal, India (dmondal_eco@yahoo.co.in)

2. * Rajib Dolai, Research Scholar, Economics, Vidyasagar University, Midnapore, West Bengal, India (rajibss44@gmail.com)

This paper examines the importance of break analysis in time series data for obtaining the best-fitted curve, using agricultural GDP data in India as a case study. Through visual assessment and statistical analysis, it is observed that the curve with breaks demonstrates a closer fit to the actual line and captures the underlying data patterns more accurately. Statistical measures, including R-squared values, Bayesian Information Criterion, average absolute residuals, and Normality test, further support the superiority of the curve with breaks in terms of explaining data variation, trade-off between fit and complexity, accuracy in prediction, and adherence to normality assumptions. Therefore, incorporating break analysis in curve fitting proves essential for achieving optimal results and improving understanding of time series patterns. Policymakers should focus on using curve-fitting models with breaks for analyzing agricultural GDP data because they perform better at capturing fluctuations and patterns, aiding informed policy decisions.

Keywords

Best-fitted Curve, Bayesian information criterion, Normality test, Curve with breaks

Bai, J., & Perron, P. (2003). Computation and analysis of multiple structural change models. Journal of Applied Econometrics, 18(1), 1-22.
Bandyopadhayay, P.S., Boik, R.J. & Vineberg, S. (1996). The curve fitting problem: A Bayesian approach. Philosophy of Science, 63(S3), S264-S272.
Bellman, R. & Roth, R. (1969). Curve fitting by segmented straight lines. Journal of the American Statistical Association, 64(327), 1079-1084.
Chernov, N., Huang, Q. & Ma, H. (2012). Does the best-fitting curve always exist? International Scholarly Research Network, 2012(1).
Dolai, R., & Mondal, D. (2023). Agricultural growth in India: A comparative analysis of sub-periods from 1960-61 to 2019-20. (2), 1-13.
Kiser, M. M. & Dolan, J. W. (2004). Selecting the best curve fit. LC GC North America, 22(2), 112-117.
Knott, G. & Shrager, R. (1972). On-line modeling by curve-fitting. ACM Siggraph Computer Graphics, 6(4), 138-151.
Mondal, D. & Saha, M.S. (2008). Growth, break and instability: Towards a unified methodology. Vidyasagar University Journal of Economics, 13, 22-36.
Wilson, J.B. (1991). Methods for fitting dominance/diversity curves. Journal of Vegetation Science, 2(1), 35-46.