# How to interpret this plot?

Please give me an explanation based on this Non-linear estimation results from the aggregate model (HOSP). This figure shows the non-linear effect of age (AGE), education (EDU), family size (FAMSZ), and log yearly household consumption expenditure (LNYHCE), respectively, on hospitalization (HOSP). Smooth function estimates and associated 95% point-wise confidence intervals in the treatment equation is obtained by applying the Joe90 copula regression spline model. The rug plot, at the bottom of each graph, shows the covariate values. The p-values for the smooth terms of AGE, EDU, FAMSZ, LNYHCE are <0.000, <0.000, <0.000 and <0.000, respectively, that is, all the variables are statistically significant at less than one percent level of significance

I don't understand how to interpret the 95% point-wise confidence bands (grey area), Joe90 copula model, the curve shape, the covariate probability and the rug plot

Information :
This is plot of the smooth function estimates for the treatment and outcome equations (and associated Bayesian intervals), for the aggregate model, the private- and the public-health insurance models, respectively. Using the Joe90 copula model, the effects of the continuous variables, age, education, family size and log yearly household consumption expenditure as a proxy for household income in the treatment and the outcome equations show different degrees of nonlinearity. The public health insurance, however, mimics the pattern observed at the aggregate level. For public health insurance, there is no clear-cut pattern in education, more so it decreases at the higher levels of education. For public health insurance, this likelihood decreases steadily with an increase in the family size. For public health insurance, the curvature is relatively flat at the lower levels of income than at the higher levels of income but rises steadily with increasing income level

• This is aggregate model for hospitalization health insurance. Using the copula regression spline model (associate with bayesian) – Rafsyaa May 23 at 23:11
• May this help you. stats.stackexchange.com/questions/101318/… – Alice May 25 at 15:35