Gfyuki

0 $begingroup$ I would like to build a supervised learning model M satisfying the following conditions: Training data {X, Y}, where x, y are numerical vectors and size(x) = m; size(y) = n Assume: M(x) = p, then: 0 < p[k] <= y[k], for all k = 1..n Could you please suggest what are the "best" loss function and optimization method that I can use to train this model M? Thank you. machine-learning optimization loss-function share | improve this question asked 13 mins ago Hoa Ngo Hoa Ngo 1 New contributor

0 $begingroup$ TO see the association of malaria prevalence with village level risk factors, I ran a Poisson model in r with the prevalence of malaria(y) as a dependent variable, altitude(x1) and Forestation(x2) as an independent variable and log of Population(x3) as an offset. Further, to estimate the independent effects of independent variables, I extracted the average marginal effects (AME) using margin command in R. res<-glmer(y~x1+x2+(1|cluster), family = poisson, offset = log(x3)) margins(res) However, now I am not quite sure how to interpret those AME numbers appropriately. You can have a look at this link which is a lead question of this question. logistic-regression share