In this work we aim to forecast tv-shares, that is the quote of tv-viewers that sees each channel. To this task, the explanatory variables (programs and date in our case) play a fundamental role. The nature of the data motivates the use of Dynamic Models. Using different formulations we show as the mentioned variables have a weak predictive content, so that often "naive" models make better than more complex formulations. Then we invoke the use of titles of the programs transmitted as explanatory variable. About this we propose a new formulation of the model that treats titles in a simple and efficient way. This proposal is implemented using a rather heuristic procedure to "mimic" titles on the basis of our data.

Dynamic Models for Non-Gaussian, Non-Linear Time Series: Forecasting Italian Tv-Shares / F. Cipollini. - STAMPA. - (1999), pp. 1-3. (Intervento presentato al convegno Second European Conference on Highly Structured Stochastic Systems tenutosi a Pavia nel September).

Dynamic Models for Non-Gaussian, Non-Linear Time Series: Forecasting Italian Tv-Shares

CIPOLLINI, FABRIZIO
1999

Abstract

In this work we aim to forecast tv-shares, that is the quote of tv-viewers that sees each channel. To this task, the explanatory variables (programs and date in our case) play a fundamental role. The nature of the data motivates the use of Dynamic Models. Using different formulations we show as the mentioned variables have a weak predictive content, so that often "naive" models make better than more complex formulations. Then we invoke the use of titles of the programs transmitted as explanatory variable. About this we propose a new formulation of the model that treats titles in a simple and efficient way. This proposal is implemented using a rather heuristic procedure to "mimic" titles on the basis of our data.
1999
Second European Conference on Highly Structured Stochastic Systems
Pavia
F. Cipollini
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/778980
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact