IEMR publishes one of the world’s largest database on mobile operator and mobile carrier metrics and forecasts.

 

Our Global Mobile Operator Forecast (GMOF) – part of our XForecast – provides operational and financial metrics for wireless markets across the world covering 800+ mobile operators in 200+ countries and territories.

 

The GMOF is the most extensive carrier-specific and country-specific forecast of its kind in the world with 2.5 million + data points and we add about 200,000 data points every  quarter.

We provide quarterly historical and quarterly forecast data starting in 1Q.2003 and ending in eight quarters beyond the current quarter. 

Mobile operator forecast

Historical Data Collection

IEMR’s Global Mobile Operator Forecast follows a rigorous methodology to capture historical data. We populate historical data in our GMOF database using proximity-based web capture technology and tracking of over 100,000 websites including operator, investor relations websites, and news websites that provide information on mobile market KPIs at the operator level.

 

We complement this data with on-line executive interviews of over 3,500 executives at Communications Service Providers (CSPs) globally. These executives are part of IEMR’s EnterpriseONE Panel. All historical data is vetted by IEMR staff to ensure accuracy and consistency.


For historical data, we also undertake a thorough panel-regression modeling exercise to estimate any missing data. Each historical predicted estimate in our dataset is based on a "best fit" panel regression model that provides an estimate of that variable for a specific operator.


Unlike other analyst firms, we do not undertake Top-down estimates based on analyst assumptions. Put together, the above data collection and estimation exercise is far superior to the typical analyst reports that are mostly based on an analyst’s estimates, not real data or data based on any rigorous modeling exercise.

 


Forecast Methodology


IEMR’s forecasts are based on a combination of time-series and structural forecasting models that are updated on a quarterly basis.

The model is a “bottom up” model that forecasts metrics at the operator-level to create country trends. For shorter term forecasts (eight quarters), we use time-series models to forecast operator-specific metrics. These country level forecasts are then aggregated up to form regional and global metrics.

SAMPLE analysis