{smcl}
{hline}
help for {hi:spell2panel}{right:(E. Leuven)}
{hline}

{title:Create panel data from time-span data}

{p 8 15}{cmdab:spell2panel} [{it:clist}] {cmd:,} {cmd:spell(}{it:date0 date1}{cmd:)} {cmd:p(}{it:period}{cmd:)} 
[{cmd:by(}{it:varlist}{cmd:)} nojoin]

{p}where {it:clist} can be a {it:varlist} or

{p}where {it:clist} can be [{cmdab:(}stat{cmdab:)}] {it:varlist} [ [{cmdab:(}stat{cmdab:)} ... ]

{p}and stat is either {cmdab:mean} or {cmdab:sum}

{title:Description}

{p}{cmd:spell2panel} creates a panel from time-span data.

{p}By default overlapping spells are broken up in disjoint adjacent spells. {it:clist} with statistics
instructs how to treat variables when spells overlap; to sum them or take an average.
 
{p}As an example, if two overlapping spells
refer to a job and the variable is hours worked one would like to sum (to get the total hours worker), 
whereas if the variable is the worker's
age one would like to take the average (which will just be the worker's age).

{p}Option {cmd:nojoin} allows the user to avoid the above procedure. Note that {cmd:spell2panel} only
generates sensical results if spells do not overlap.

{title:Options}

{p 0 4} {cmd:spell(}{it:date0 date1}{cmd:)} start and end date of spell.

{p 0 4} {cmd:p(}{it:period}{cmd:)} define aggregation period:
y,h,q,m,w,d. That is: yearly, halfyearly, quarterly, monthly, weekly, daily.

{p 0 4} {cmd:by(}{it:varlist}{cmd:)} defines the groups (id's) for which the panel must be constructed.

{title:Example}

{title:Acknowledgement}

{p}Thanks to Barbara Sianesi for suggesting the clever use of expand and pointing me to Stata's time series formats.

{title:Author}

{p}E. Leuven, Department of Economics, University of Amsterdam. If you observe any problems {browse "mailto:leuven@fee.uva.nl"}.