Normalised MSE of the PMH, NPMC and ABC-SMC algorithms versus running time.

<p>The figure displays the average normalised MSE per parameter attained by the NPMC scheme (with <i>M</i> = 50, 200 and 400 samples per iteration, after <i>K</i> = 20 iterations), the PMH algorithm with chains of length <i>L</i> = 1,000, <i>L</i> = 4,000 and <i>L</i> = 8,000, and scale parameter <i>s</i>² = 0.1, and the ABC-SMC method with <i>J</i> = 200 and <i>J</i> = 800 samples per population. The lowest error is achieved by the NPMC algorithm, which demands a running time slightly shorter than the PMH schemes and much shorter than the (average) running time of the ABC-SMC method. Time is given in minutes.