poor strategy. It is immediately seen that e.g. s6 is better, because its expected losses are smaller, whatever the state of nature is. This is expressed by saying that S3 is dominated by s6. A strategy s dominates s* if L(0i,s) L(0i,s*) and 1.(02,s) 1.(02,s*) but the 130 iMinimax expected loss TllMlnimax risk R(eus) Figure 2. .Minimax risk Figure 3.

Digitale Tijdschriftenarchief Stichting De Hollandse Cirkel en Geo Informatie Nederland

Tijdschrift voor Kadaster en Landmeetkunde (KenL) | 1967 | | pagina 12