Understanding Society User Support: Issueshttps://iserredex.essex.ac.uk/support/https://iserredex.essex.ac.uk/support/support/favicon.ico?15995719382015-04-21T08:22:54ZUnderstanding Society User Support
Redmine Understanding Society User Support - Support #359 (Closed): savings greater than incomehttps://iserredex.essex.ac.uk/support/issues/3592015-04-21T08:22:54ZElisa Macchialterelim@gmail.com
<p>Dear all,</p>
<p>working on the BHPS I noticed that there is a lot of observation which show annual saving (amount saved each month * 12) is greater than total annual income (fiyr). Most of these people are originally from the U.K. ( so remittance can be excluded) and not all of them are married. Such problem is present even if I reduce the sample to people older than 40.</p>
<p>Do you have an explanation for this evidence? I assumed that "fiyr" contained all the possible sources of income.</p>
<p>Kind regard,</p>
<p>Elisa</p> Understanding Society User Support - Support #357 (Closed): cluster variablehttps://iserredex.essex.ac.uk/support/issues/3572015-04-11T13:49:43ZElisa Macchialterelim@gmail.com
<p>I am working on the BHPS dataset. My aim is to check whether the change in the Long Term Care policies in 2002 had an effect on precautionary savings at an individual level.<br />To do so, I need to use the BHPS as a panel: I saw that there are longitudinal weights that are to be used in order to make longitudinal studies on the data. In addition to that, do you think I should use also cluster and strata variables when declaring the dataset? I understood the dataset is has a clustered design but as far as I saw, such cluster (PSU) and strata variables are contained in the HHsamp but not in the Indresp. So I thought that maybe, if the analysis is done at an individual level the longitudinal weights are already taking everything into account.</p>
<p>kind regards, <br />Elisa Macchi</p>