On Sat, 27 Feb 2016, Marc O'Callaghan wrote: > Dear all, > > I have a time series, which comprises 31 observations per year (actually > the observations for March, set in a time series with periodicity 31). > My wish is to compact it to the daily average, i.e. end up with 31 > observations, each of which represents the average value of that date > throughout the available years. I therefore sought to compact it with > "dataset compact 31", to which I was told "This conversion is not > supported". I then tried other conversions, but it appears no expansion > or compaction is accepted at all. > > I assume this is due to the unorthodox periodicity, but my question is > then: how can I successfully compact my data into 31 average values ?
The simplest way would be, I think, to redefine your dataset as a panel dataset, with the day as unit variable and the year as time bariable and then use the pmean() function. Or otherwise, you could use the aggregate() function to create a matrix that you re-use as the starting point of your new dataset. HTH, ------------------------------------------------------- Riccardo (Jack) Lucchetti Dipartimento di Scienze Economiche e Sociali (DiSES) Università Politecnica delle Marche (formerly known as Università di Ancona) r.lucchetti(a)univpm.it http://www2.econ.univpm.it/servizi/hpp/lucchetti -------------------------------------------------------
