Stepwise selection of functional covariates in forecasting peak levels of olive pollen Escabias Machuca, Manuel Valderrama Bonnet, Mariano José Aguilera Del Pino, Ana María Santofimia, M. Elena Aguilera Morillo, María del Carmen Olea europaea L. airborne pollen Functional-logit-regression Selection of functional predictors High levels of airborne olive pollen represent a problem for a large proportion of the population because of the many allergies it causes. Many attempts have been made to forecast the concentration of airborne olive pollen, using methods such as time series, linear regression, neural networks, a combination of fuzzy systems and neural networks, and functional models. This paper presents a functional logistic regression model used to study the relationship between olive pollen concentration and different climatic factors, and on this basis to predict the probability of high (and possibly extreme) levels of airborne pollen, selecting the best subset of functional climatic variables by means of a stepwise method based on the conditional likelihood ratio test. 2022-02-23T12:13:45Z 2022-02-23T12:13:45Z 2012-10-27 info:eu-repo/semantics/article Escabias, Manuel & Valderrama, Mariano & Aguilera, Ana & Santofimia, M. & Aguilera-Morillo, M.. (2012). Stepwise selection of functional covariates in forecasting peak levels of olive pollen. Stochastic Environmental Research and Risk Assessment. 27. 10.1007/s00477-012-0655-0. http://hdl.handle.net/10481/72974 27. 10.1007/s00477-012-0655-0 eng http://creativecommons.org/licenses/by-nd/3.0/es/ info:eu-repo/semantics/embargoedAccess Atribución-SinDerivadas 3.0 España Springer