%PDF-1.4 % 5 0 obj << /S /GoTo /D (section.1) >> endobj 8 0 obj (Introduction) endobj 9 0 obj << /S /GoTo /D (section.2) >> endobj 12 0 obj (Background) endobj 13 0 obj << /S /GoTo /D (subsection.2.1) >> endobj 16 0 obj (Notation) endobj 17 0 obj << /S /GoTo /D (subsection.2.2) >> endobj 20 0 obj (Quantities of Interest) endobj 21 0 obj << /S /GoTo /D (subsection.2.3) >> endobj 24 0 obj (Existing matching methods and practice) endobj 25 0 obj << /S /GoTo /D (section.3) >> endobj 28 0 obj (Coarsened Exact Matching) endobj 29 0 obj << /S /GoTo /D (subsection.3.1) >> endobj 32 0 obj (The Algorithm) endobj 33 0 obj << /S /GoTo /D (subsection.3.2) >> endobj 36 0 obj (The Benefits) endobj 37 0 obj << /S /GoTo /D (section.4) >> endobj 40 0 obj (An Extended Example) endobj 41 0 obj << /S /GoTo /D (subsection.4.1) >> endobj 44 0 obj (Basic Evaluation and Analysis of Unmatched Data) endobj 45 0 obj << /S /GoTo /D (subsection.4.2) >> endobj 48 0 obj (Coarsened Exact Matching) endobj 49 0 obj << /S /GoTo /D (subsubsection.4.2.1) >> endobj 52 0 obj (Automated Coarsening) endobj 53 0 obj << /S /GoTo /D (subsubsection.4.2.2) >> endobj 56 0 obj (Coarsening by Explicit User Choice) endobj 57 0 obj << /S /GoTo /D (subsubsection.4.2.3) >> endobj 60 0 obj (Coarsening categorical variables) endobj 61 0 obj << /S /GoTo /D (subsection.4.3) >> endobj 64 0 obj (Restricting the matching solution to a k-to-k match) endobj 65 0 obj << /S /GoTo /D (subsection.4.4) >> endobj 68 0 obj (Estimating the Causal Effect from cem output) endobj 69 0 obj << /S /GoTo /D (subsection.4.5) >> endobj 72 0 obj (Matching and Missing Data) endobj 73 0 obj << /S /GoTo /D (subsubsection.4.5.1) >> endobj 76 0 obj (Matching on Missingness) endobj 77 0 obj << /S /GoTo /D (subsubsection.4.5.2) >> endobj 80 0 obj (Matching Multiply Imputed Data) endobj 81 0 obj << /S /GoTo /D (subsection.4.6) >> endobj 84 0 obj (Blocking in Randomized Experiments) endobj 85 0 obj << /S /GoTo /D (subsection.4.7) >> endobj 88 0 obj (Using cem to Improve Other Matching Methods) endobj 89 0 obj << /S /GoTo /D (section.5) >> endobj 92 0 obj (cem -- Coarsened Exact Matching) endobj 93 0 obj << /S /GoTo /D (subsection.5.1) >> endobj 96 0 obj (Syntax) endobj 97 0 obj << /S /GoTo /D (subsection.5.2) >> endobj 100 0 obj (Description) endobj 101 0 obj << /S /GoTo /D (subsection.5.3) >> endobj 104 0 obj (Options) endobj 105 0 obj << /S /GoTo /D (subsection.5.4) >> endobj 108 0 obj (Output) endobj 109 0 obj << /S /GoTo /D (section.6) >> endobj 112 0 obj (balance - \(Im\)balance Measures for CEM) endobj 113 0 obj << /S /GoTo /D (subsection.6.1) >> endobj 116 0 obj (Syntax) endobj 117 0 obj << /S /GoTo /D (subsection.6.2) >> endobj 120 0 obj (Description) endobj 121 0 obj << /S /GoTo /D (subsection.6.3) >> endobj 124 0 obj (Arguments) endobj 125 0 obj << /S /GoTo /D (subsection.6.4) >> endobj 128 0 obj (Saved Results) endobj 129 0 obj << /S /GoTo /D [130 0 R /Fit ] >> endobj 132 0 obj << /Length 884 /Filter /FlateDecode >> stream xڵVMo6WHoRK4ɺE-eWbmɐd;i@/&9̛G)nNȑb~rvxA5fTb,17bMjCY egnQV0t6!MomhJЃoŷi6KZJcUR`C +-S, Fߞ[2,456pV̹It*qmb~w-mۥP3l xe0$&o!}Ch~n)XS,PD]c5NAL5-*.l\!\4b4Ed靈k]( sW1$qh:k6@nB*4skے68.w"oWx6jҮは_N G?`"C,siހVT O(4_*ᜣ_(a_Ro((ߞ|i w`ؾeKfv&5X+Љ~aR! ILUӵ7ClEvk>H"}\f?kZRTV%+fަˮ\G"(]ѧiUФLdv,5]I@`߆mÝ2A%&j4@eAYd-h#5c9` ʢ kQD(s, }%Ȟ({oxa&>cOtZ{x.2rQTGͥDD-٘>B0uȿO*:Iƽ 2&c12hI콃'>E^zlw4;SA8.cc KweO_87v QôHG\OpE endstream endobj 130 0 obj << /Type /Page /Contents 132 0 R /Resources 131 0 R /MediaBox [0 0 612 792] /Parent 141 0 R >> endobj 133 0 obj << /D [130 0 R /XYZ 99.213 780.045 null] >> endobj 134 0 obj << /D [130 0 R /XYZ 99.213 748.864 null] >> endobj 131 0 obj << /Font << /F23 135 0 R /F24 136 0 R /F25 137 0 R /F7 138 0 R /F8 139 0 R /F50 140 0 R >> /ProcSet [ /PDF /Text ] >> endobj 144 0 obj << /Length 1225 /Filter /FlateDecode >> stream x}VK6ϯ0v0[ۃb3P*ɓߗ9ɤH(>~t8)ӗE$qTulAZUQAQQ ?3qZ