In　this　paper,　we　study　the　estimation　of　the　partial　linear　spatial　autoregressive　model　with　the　response　variable　missing　at　random.　Based　on　the　nonparametric　B-spline　method,　the　marginal　maximum　Likelihood　estimation　and　the　EM　algorithms　for　the　maximum　Likelihood　estimation　and　the　pseudo-restricted　maximum　Likelihood　estimation　are　proposed,　respectively.　Numerical　simulations　are　carried　out　under　different　sample　sizes,　missing　rates　and　spatial　weight　matrix　settings　to　compare　the　performances　of　the　three　methods.　Finally,　the　effectiveness　of　the　three　estimation　methods　are　verified　by　a　real　data　analysis.　©　2023　Chinese　Academy　of　Sciences.　All　rights　reserved.