Skip to main content

Table 3 Monte Carlo experiment for four practices and 1,000 observations (10,000 draws)

From: A note on testing for complementarity and substitutability in the case of multiple practices

True effect σε = 0.25 σε = 1 σε = 3.5
  Compl Neither Subst Compl Neither Subst Compl Neither Subst
Multiple-restrictions test
 Complements 3,106 72 0 3,014 160 4 1,942 1,217 19
 Neither 0 3,592 2 350 2,860 384 634 2,320 640
 Substitutes 0 88 3,140 3 185 3,040 13 1,287 1,928
 Correct (%)   98.38%    89.14%    61.90%  
All cross-term test
 Complements 3,172 6 0 3,069 108 1 2,235 933 10
 Neither 1,510 515 1,569 1,160 1,221 1,213 540 2,536 518
 Substitutes 0 16 3,212 1 128 3,099 7 1,033 2,188
 Correct (%)   68.99%    73.89%    69.59%  
Single cross-term test
 Complements 2,870 149 159 2,806 253 119 2,326 795 57
 Neither 1,294 938 1,362 1,070 1,406 1,118 525 2,559 510
 Substitutes 176 188 2,864 135 298 2,795 61 956 2,211
 Correct (%)   66.72%    70.07%    70.96%