library(plotly) Prediction_map_3D = Prediction_map[!is.na( Prediction_map$Prediction), ] plot_ly(Prediction_map_3D, x = ~X, y = ~Y, z = ~Prediction, type = "scatter3d", mode = "markers", marker = list(size = 3, opacity = 0.8, color = ~Prediction, colorscale = "Viridis")) %>% layout(scene = list(xaxis = list(title = "Latitude", showgrid = TRUE, gridwidth = 10), yaxis = list(title = "Longitude", showgrid = TRUE, gridwidth = 10), zaxis = list(title = "d18O"),tickvals = rev(seq(min(Prediction_map_3D$Prediction), max(Prediction_map_3D$Prediction), length.out = 5)),camera = list(eye = list(x = 0, y = -0.1, z = 2))), showlegend = FALSE)

Avaliar o monitoramento de animais domésticos no Campus Marco Zero da Universidade Federal do Amapá.

Daniel Koiti Oshiro e Laura Lucia Dominguez

This report explores the concepts of Bayesian regression as detailed in Chapter 4 of "Statistical Rethinking" by Richard McElreath. Through practical examples and visualizations, it compares Bayesian and frequentist approaches to linear regression, demonstrating how Bayesian methods provide a more nuanced understanding of parameter uncertainty by using probability distributions. The report also delves into the natural emergence of normal distributions from the addition of random variables and justifies the use of Gaussian distributions in statistical modeling due to their ontological and epistemological foundations.