| Abstract [eng] |
Lately the modelling of innovation process is a constantly recurring problem. As with any new service for managers it is important to understand and to able to forecast the market developement for strategic, capacity and financial planning purposes. This master thesis analyses computer penetration data in 23 different countries and tries to propose the best fitting model for forcasting. Penetration rate usually shows the popularity of a brand or a product. It is given as a percentage of the countries’ households who have bought that particular brand or product at least once within a defined period of time. In marketing there is a decision to make: enlarge sales by trying to attract more costumers in the sales region or by expanding the population of potential buyers, i.e. by expanding to other emerging markets. In the beginning of this thesis we will introduce some basic information about the panel data and some approaches of the theory needed for estimation. Later on we will pass to the main part of the work which is devoted to choosing the model. At first, we analyze some linear trend models with autoregression terms and exogenous variables. Then we examine a diffusion process by approximating it with a quadratic trend, logistic and Gompertz models. Also some basic information about the Bass diffusion model will be provided. |
