Case studies - 10 August 2023

Accuracy of annual forecasts for the pay TV and VOD services market

Accuracy of annual forecasts for the cosmetics retail market

We determined the forecast accuracy by using the mean absolute percentage error (MAPE), demonstrating how much PMR’s forecasts differed, on average, from the actual values: 

 

 

where n is the number of periods (years) for which the variable is forecast. 

 

The crucial obstacle to the proper verification of the forecast accuracy was a partial change in historical data prompted by the inclusion of new information and the revision of some market size and value estimates. These changes are not particularly relevant to the market picture but affect the forecast accuracy and make it difficult to calculate the MAPE. 

 

We therefore present MAPEs which are determined in two ways. The first, unadjusted, MAPE, is based directly on data and forecasts from our 2016-2023 reports and current data. These errors are potentially magnified by changes in historical data. The degree of magnitude of the error caused by the changes in historical data is also presented below.  

 

In addition, we present the adjusted MAPE, after eliminating the impact of the historical data changes. Changes in the data mainly affected the values and, to a lesser extent, the rate-of-change, of the forecast indicators. It is, therefore, reasonable to adjust the old historical data to match the new data by adopting the same values for the historical data up to the start of the forecasts and then calculating the forecast values by using the projected change, to compare them with the actual values.  

 

With this approach, the adjusted MAPEs are below 5% and usually fall substantially short of 3%. In addition, in the absence of historical data adjustment, errors greater than 5% are rarely recorded, and in the more recent 2018-2022 forecasts they are, accordingly, less significant, customarily falling short of 3%. From a technical, i.e. statistical, viewpoint, this demonstrates that PMR’s forecasts are accurate. 

 

The predictive accuracy assessment was based on data from our report entitled: “Pay TV and VOD market in Poland 2022” and its previous versions. 

 

The table below summarises the MAPE unadjusted, mean historical data adjustment error and MAPE adjusted for the forecasts pertaining to the number of pay TV subscribers and the pay TV market value in 2016-2022.  

ReportAverage MAPE for: Unadjusted MAPE Mean historical data error Adjusted MAPE
Number of pay TV subscribers (million) in Poland 20162016–2021 5,5%2,5%0,5%
Number of pay TV subscribers (million) in Poland 20172017–2022 3,1%1,1%1,1%
Number of pay TV subscribers (million) in Poland 20182018–2022 1,2%1,0%0,7%
Number of pay TV subscribers (million) in Poland 20192019–2022 0,8%1,0%1,8%
Number of pay TV subscribers (million) in Poland 20202020–2022 1,2%0,4%0,3%
Number of pay TV subscribers (million) in Poland 20212021–2022 2,0%0,2%1,4%
Number of pay TV subscribers (million) in Poland 202220220,9%0,2%0,2%
Pay TV market value (PLN bn) in Poland 20162016–2021 8,6%5,7%3,0%
Pay TV market value (PLN bn) in Poland 20172017–2021 5,3%0,5%5,1%
Pay TV market value (PLN bn) in Poland 20182018–2021 6,9%0,7%4,7%
Pay TV market value (PLN bn) in Poland 20192019–2021 1,1%0,4%1,1%
Pay TV market value (PLN bn) in Poland 20202020–2021 3,3%0,3%4,9%
Pay TV market value (PLN bn) in Poland 20212021–2021 1,4%0,1%1,1%
Pay TV market value (PLN bn) in Poland 202220222,0%0,1%1,5%
Do you need more data?

In PMR, we advise and assist in the operations of your company

Maciej Gazda

Head of Forecasting