International Journal of Engineering Research in Electronics and Communication Engineering

Financial Crisis Prediction using a Hybrid ANN Model

Author : Ranjana Pagui ¹

Date of Publication :20th April 2017

Abstract: Several economic crises have already been witnessed by the world in the past few decades. This study proposes a hybrid model combining Artificial Neural Networks (ANN) and K Nearest Neighbour (KNN) models serially in predicting financial crisis. The analysis is done using dataset of the past 15 years comprising of 13 macroeconomic indicators of the country. The macroeconomic indicators are chosen based on their relevance to the economic conditions. The crucial macroeconomic indicators used in this study include consumer price index, export prices, import prices, terms of trade, foreign direct investment, government spending, producer prices, inflation rate, unemployment rate, GDP, money supply(m1), GDP/capita and industrial production. The prediction results obtained are compared separately with the results of ANN and KNN. Overall, the hybrid model possesses a higher prognostic ability and the likeliness of financial crisis in India during 2016 and 2017 is low

Reference :

1. Zhou Jie, Lin Yan, Liu Xin, “Research on financial crisis prediction model based on Rough Sets and Neural Network”, IEEE, (2011)
2. Zhang Xi-shua, “Constructing the Model of Enterprise Financial Crisis with Principal Component Analysis and ANN”, IEEE (2010).
3. Mushang Lee1, Ching-Hui Shih, Tsui Chih Wu, “The emerging financial pre-warning systems”, IEEE (2012).
4. Liu Hengjun, “Rough Set Neural Network based financial distress Prediction”, IEEE (2014)
5. Sankha Pallab Saha, “On Small Sample Prediction of Financial Crisis”, IEEE (2009).