Digital media ADEX forecast in India involving machine learning and statistical based modelling techniques

Abstract

The popularity of digital advertising has gone through a sea change in the recent past owing to growing adoption of smart phones and the effect of the recent pandemic due to which consumers confined at home, spent more time on digital media. Advertising expenditure (ADEX) is considered as the most appropriate variable that represents adoption of advertising media. This paper aims to assess the adoption of the digital media in future, through predictive analysis, using a best fit model. The study has used univariate time series data and both statistical and machine learning approaches have been considered. Iterative techniques yield multiple models and accuracy estimates determine the quality of the best fitted model. Neural network auto regressive model is found to best fit the historical data for the purpose of projection. The study outcomes indicate mobile and video to have highest adoption followed by social media in terms of CAGR from 2018 till 2022. However, by 2022, it is expected that search ads will continue to attract highest advertising investment closely followed by mobile ads.