Forecasting and analysing the characteristics of 3G and 4G mobile broadband diffusion in India: A comparative evaluation of Bass, Norton-Bass, Gompertz, and logistic growth models
Forecasting and analysing the characteristics of 3G and 4G mobile broadband diffusion in India: A comparative evaluation of Bass, Norton-Bass, Gompertz, and logistic growth models
复制标题
DOI：
10.1016/j.techfore.2019.119885
复制
发表时间：
2020-03-01
影响因子：
12
通讯作者：
Saha, Debashis
中科院分区：
文献类型：
--
作者：
Jha, Ashutosh;Saha, Debashis
对第三代（3G）或/和第四代（4G）移动的宽带服务（MBS）在全国范围内的传播的经验性理解已被证明对该国的业务和政策具有影响。然而，现存的文献缺乏解释，这些服务在印度的扩散和预测。我们通过分析印度3G和4G服务的个人和多代扩散来解决这一差距，使用Bass，Gompertz，Logistic和Norton-Bass模型，这些模型利用线性和非线性回归技术的混合。此外，我们评估了几个外生变量对这些MBS在印度的扩散的影响。我们的分析表明，首先在扩散的情况下，Bass模型估计对3G和4G历史数据都非常敏感，而Gompertz和Logistic模型与相同的数据集拟合良好。正如预期的那样，Norton-Bass模型-包括所有连续几代的2G，3G和4G -提供了更可靠的扩散参数估计。其次，就3G预测而言，Bass模型更适合于对最终市场潜力的固定假设，而Gompertz和Logistic模型似乎更适合于“乐观”的长期预测和“保守”的短期预测。我们的结果还显示，4G在印度的扩散速度是3G扩散速度的6.1倍，到2026年，印度MBS订阅总量可能达到4.1亿。最后，我们注意到，在显着的外部影响力来源，2012年国家电信政策，每用户平均收入，总收入变量有显着的积极影响的MBS在印度的扩散。
An empirical understanding of countrywide diffusion of third (3G) or/and fourth (4G) generations of Mobile Broadband Services (MBSs) has proven implications for both business and policy of the country. However, extant literature lacks in explanation for the diffusion and forecast of these services in India. We address this gap by analyzing both individual and multigenerational diffusions of 3G and 4G services in India, using Bass, Gompertz, Logistic and Norton-Bass models that utilize a mix of linear and non-linear regression techniques. Additionally, we evaluate the influence of several exogenous variables on the diffusion of those MBSs in India. Our analyses reveal that, firstly in case of diffusion, Bass model estimates are quite sensitive to both 3G and 4G historical data, whereas Gompertz and Logistic models fit well with the same dataset. As expected, Norton-Bass model - encompassing all the successive generations of 2G, 3G and 4G - provides more reliable estimates of the diffusion parameters. Secondly, as far as 3G forecast is concerned, Bass model works better with fixed assumptions of ultimate market potential, whereas Gompertz and Logistic models seem to be more suited for 'optimistic' long-range forecasts and 'conservative' short-term forecasts, respectively. Our results also show that 4G is diffusing at 6.1 times the speed of 3G diffusion in India, when the total MBS subscription in India is likely to reach 410 million by 2026. Finally, we notice that, among the notable external sources of influence, National Telecom Policy 2012, average revenue per user, and aggregate income variables have significant positive impacts on the diffusion of MBSs in India.