Survey on During Pandemic scenario: Domestic Retail Sales Forecasting of Passenger Vehicles in India using Time Delay Neural Network

Main Article Content

Dr. Vijay Bhardwaj, Dr. Sunny Arora

Abstract

Accurate sales forecast is crucial to plan the production process, especially in automotive
industry, where a large number of factors –macro-economic, micro-economic and others
dynamically influence the demand.Forecasting in such a dynamic industry, especially with the
onset of COVID, has become extremely important with the preference for personal
transportation on the increase. In this paper, a neural network model to forecast the domestic
retail sales of passenger vehicles in India is formulated. A Time Delay Feed Forward Neural
Network with Back Propagation (TDNN)is used and the forecasting accuracies determined by
comparing the values of Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean
Absolute Deviation (MAD) and Mean Absolute Percentage Error (MAPE). The macroeconomic
indicators identified were Unemployment rate, GDP growth rate, Private Final
Consumption Expenditure (PFCE), Index of Industrial Production (IIP), Bank Lending Rate
(BLR), Inflation Rate and Crude oil prices. The indicators were taken as the input variables and
Passenger Vehicle Retail Sales was selected as the target variable to formulate the TDNN
model. It was observed that the TDNN model was able to forecast the output with great
accuracy.

Article Details

Section
Articles