{smcl} {* *! version 4.0 10july2011}{...} {cmd:help spdir} {hline} {title:Title} {title:spdir - Creates {it}directed dyad{bf} contagion spatial effect variable} {title:Syntax} {p 8 17 2} {cmdab:spdir} lagvar {ifin}, {opt w:eightvar(varname)} {opt s:ource(varname)} {opt t:arget(varname)} {opt link(options)} [{cmd:}{it:options}] {synoptset 20 tabbed}{...} {synopthdr} {synoptline} {synopt:{opt time(varname)}}contains the numeric time variable{p_end} {synopt:{opt excl:usive}}exclusive directed dyad contagion{p_end} {synopt:{opt nor:owst}}spatial effect variable not row-standardized{p_end} {synopt:{opt nom:erge}}no automatic merge of spatial effect variable into original dataset{p_end} {synopt:{opt se:name(name)}}name to be given to created spatial effect variable{p_end} {synopt:{opt label:name(name)}}name of label given to spatial effect variable{p_end} {synopt:{opt file:name(name)}}name of file to which spatial effect variable saved{p_end} {synoptline} {p2colreset}{...} {title:Description} {pstd} {cmd: spdir} generates a directed dyad contagion spatial effect variable for analysis of spatial dependence in directed dyad data. It can create spatial effect variables for spatial lag, spatial-x and spatial error models. See Neumayer and Plümper (2010) for an exposition of all possible forms of contagion in directed dyad data. See Plümper and Neumayer (2010) for a discussion of model specification in the analysis of spatial dependence. {title:Additional information} {pstd} See {browse "http://personal.lse.ac.uk/neumayer/spdir.htm"}. {title:Arguments} {phang} {opt lagvar} is the variable to be spatially lagged. It is the directed dyadic dependent variable in spatial lag models, a selected independent variable in spatial-x models and a saved regression residual in spatial error models. {phang} {opt w:eightvar(varname)} is the weighting or connectivity variable linking source unit {it}i{sf} with target unit {it}j{sf}. It must be numeric and must not contain negative values. It may or may not be directed. {phang} {opt s:ource(varname)} is the identifying variable of source unit {it}i{sf}. It can be a numeric or string variable. {phang} {opt t:arget(varname)} is the identifying variable of target unit {it}j{sf}. It can be a numeric or string variable. {title:Options} {phang} {opt link(options)} is required. The following options are allowed: {it}ik{sf}, {it}ki{sf}, {it}im{sf}, {it}mi{sf}, {it}jm{sf}, {it}mj{sf}, {it}jk{sf}, {it}kj{sf}. Option {it}ik{sf} requests that the virtually transformed weighting variable {opt w:eightvar(varname)} is to represent connectivity from source unit {it}i{sf} to other source units {it}k{sf}. Option {it}ki{sf} requests connectivity from other source units {it}k{sf} to source unit {it}i{sf}. Option {it}im{sf} requests connectivity from source unit {it}i{sf} to target units {it}m{sf}. Option {it}mi{sf} requests connectivity from target units {it}m{sf} to source unit {it}i{sf}. Option {it}jm{sf} requests connectivity from target unit {it}j{sf} to other target units {it}m{sf}. Option {it}mj{sf} requests connectivity from other target units {it}m{sf} to target unit {it}j{sf}. Option {it}jk{sf} requests connectivity from target unit {it}j{sf} to source units {it}k{sf}. Option {it}kj{sf} requests connectivity from source units {it}m{sf} to target unit {it}j{sf}. Which of these options is appropriate depends on the specific hypothesis of spatial dependence to be tested. Researchers can request the sum or product of any two of these connectivities by listing the first one, then either + or * and then the second link. For example, option {it}ik+jm{sf} requests that the weighting matrix represents the {it}sum{sf} of connectivities invoked by {it}ik{sf} and {it}jm{sf}. Option {it}ik*kj{sf} requests that the weighting matrix represents the product of connectivities invoked by {it}ik{sf} and {it}kj{sf}. The order does not matter, i.e., option {it}ik+jm{sf} is the same as option {it}jm+ik{sf}. The same links cannot be combined with each other, i.e., {it}ik+ik{sf} is not an allowed option. {phang} {opt time(varname)} is an optional argument. If users wish to generate a time-varying spatial effect variable, then the numeric time variable must be stated here. {phang} {opt excl:usive} specifies that all dyads containing either {it}i{sf} or {it}j{sf} as either source or target are excluded from having a spatial effect on dyad {it}ij{sf}. {phang} {opt nor:owst} requests that the generated spatial effect variable is not row-standardized. See Plümper and Neumayer (2010) for an explanation and discussion of row-standardization. Row-standardization is the default option. {phang} {opt nom:erge} requests that the generated spatial effect variable is not automatically merged into the original data set. {phang} {opt se:name(name)} names the generated spatial effect variable. In the default option, if the weighting matrix is row-standardized, then the generated spatial effect variable is called SE_var_dirdyad_rowst. If the weighting matrix is not row-standardized, then the spatial effect variable is called SE_var_dirdyad_norowst. Any previously existing variable with the same name will be replaced. {phang} {opt label:name(name)} names the label of the generated spatial effect variable. The default label given is "Directed dyad contagion spatial effect variable". {phang} {opt file:name(name)} requests that a dataset containing the generated spatial effect variable is saved in the current working directory under the defined name. In the default option, if the weighting matrix is row-standardized, then a file is saved in the current working directory called SE_file_dirdyad_rowst. If the weighting matrix is not row-standardized, then the saved file is called SE_file_dirdyad_norowst. Any previously existing file with the same name will be replaced. {title:References} {pstd} Neumayer, Eric and Plümper, Thomas. 2010. Spatial Effects in Dyadic Data. {it}International Organization{sf} 64 (1), pp. 145-165. {pstd} Plümper, Thomas and Eric Neumayer. 2010. Model Specification in the Analysis of Spatial Dependence. {it}European Journal of Political Research{sf} 49 (3), pp. 418-442. {title:Examples} {phang}{cmd:. spdir y, w(exports) s(source_country) t(target_country) link(ik) sename(se_dirdyad) filename(dirdyad_file)} {phang}{cmd:. spdir y, w(exports) s(source_country) t(target_country) time(year) link(jm) norowst} {phang}{cmd:. spdir y, w(exports) s(source_country) t(target_country) time(year) link(ik*jm) nomerge} {title:Authors} {pstd} Eric Neumayer{p_end} {pstd} Department of Geography and Environment{p_end} {pstd} London School of Economics and Political Science (LSE){p_end} {pstd} London WC2A 2AE, UK{p_end} {pstd} e.neumayer@lse.ac.uk{p_end} {pstd} {browse "http://personal.lse.ac.uk/neumayer"}{p_end} {pstd} Thomas Plümper{p_end} {pstd} Department of Government{p_end} {pstd} University of Essex{p_end} {pstd} Wivenhoe Park{p_end} {pstd} Colchester CO4 3SQ, UK{p_end} {pstd} tpluem@essex.ac.uk{p_end} {pstd} {browse "http://www.polsci.org/pluemper/"}{p_end}