
The Continuous Case: Gaussian Bayesian Networks 57
As we can see from the code above, the evidence for this method is provided
by a list of values, one for each conditioning variable. Similarly, the probability
of a specific event can be computed using likelihoo d weighting via cpquery.
So, for example, we may be interested in the probability of having a vegetative
mass above 70 in the conditions specified by G and E.
> cpquery(gbn.bnlearn, event = (V > 70),
+ evidence = list(G = 10, E = 90), method = "lw")
[1] 0.00935
The probability we obtain is very low despite the favourable environments, as
expected from such bad genotypes.
2.7 Plotting Gaussian Ba ...