Look for slope dummies or differential slope dummies in the index to your econometric texts. There is some material in Gujarati and Porter (2009) which iss the first text that I checked. The basic theory is that your X'X matrix must be non singular. Thus you can add one of LogAvgTemp x Spring or LogAvgTemp. to your X matrix. Both are equivalent but one may be more convenient than the other. (Running both may save you some calculations and/or make some test statistics easier to calculate). I hope that this is of some help. John On 13 May 2011 13:04, Johannes Lips <johannes.lips(a)googlemail.com&gt; wrote: > Hello, > > sorry for posting such a probably off-topic question to this list but I > don't know of any other place I could ask. > > I am fitting an arima model with exogenous variables and was trying to > identify if there exist differences between the relationship depending on > the season of the year. > Therefore I added a combined Variable which combines a binary dummy for the > season (Summer, Fall, Winter) with the exogenous variable the logarithmic > average temperature on a particular day. So I have three Dummies (LogAvgTemp > x Summer, ..., LogAvgTemp x Winter). > Now I don't know if it's sufficient if I add those three combined dummies or > if I should also add the LogAvgTemp as an additional exogenous variable. > Perhaps someone could point me to a text book or something similar which > might address such a problem. > > Thanks in advance and once again sorry for being slightly off-topic, > > johannes > > _______________________________________________ > Gretl-users mailing list > Gretl-users(a)lists.wfu.edu > http://lists.wfu.edu/mailman/listinfo/gretl-users > -- John C Frain Economics Department Trinity College Dublin Dublin 2 Ireland www.tcd.ie/Economics/staff/frainj/home.html mailto:frainj@tcd.ie mailto:frainj@gmail.com _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users