{smcl}
{* *! version 1.0.0 07oct2019 Matteo Bottai}{...}

{cmd:help m_power}
{hline}

{hi:m_power}: Power and sample size to detect slope differences between two groups with mixed effects models

{title:Syntax}

{p 8 13 2}
{cmd:m_power [, {it:options}]}

{title:Description}

{pstd}
{cmd:m_power} estimates the detectable difference in the slope of a longitudinal linear trend between two groups using a random-intercept, random-slope mixed effect model (Yi and Panzarella, 2002).

{title:Options}

{synoptset 13 tabbed}{...}
{synopt:{opt diff:erence}}specifies the detectable difference; cannot be used if subjects is specified{p_end}
{synopt:{opt s:ubjects}}specifies the number of subjects; cannot be used if difference is specified{p_end}
{synopt:{opt t:imepoints}}specifies the number of timepoints{p_end}
{synopt:{opt level}}specifies the level of the test{p_end}
{synopt:{opt power}}specifies the desired power{p_end}
{synopt:{opt sigma}}specifies the residual scalar standard deviation{p_end}
{synopt:{opt r}}specifies the residual correlation matrix in mata syntax{p_end}
{synopt:{opt d}}specifies the random coefficient covariance matrix in mata syntax{p_end}

{title:Examples}

{phang}{stata "m_power, subjects(100)"}{p_end}
{phang}{stata "m_power, difference(1)"}{p_end}
{phang}{stata "m_power, d(2,0\0,.5) sigma(1.5) subjects(200) power(.9)"}{p_end}

{title:Also see}

{phang}{manhelp mixed XT:mixed}, {manhelp power R:power}{p_end}

{title:Reference}

{pstd}Yi Q. and Panzarella T (2002). Estimating sample size for tests on trends across repeated
measurements with missing data based on the interaction term in a mixed model. Controlled Clinical Trials 23: 481-496.{p_end}

{title:Author}

{pstd}Matteo Bottai{p_end}
{pstd}Unit of Biostatistics{p_end}
{pstd}{browse "http://ki.se/":Institute of Environmental Medicine, Karolinska Institutet}{p_end}
{pstd}Stockholm, Sweden{p_end}