Hello. I would like to ask a question, are there any test statistics to possibly use for underdispersed count data aside from generalized poisson model?
Hi, when you do the scatter plot between Ln(y) and X, which Xp are you considering? If I am not mistaken when you have more than one covariate you can´t graph the multiple linear regression
The graph for simplicity shows one variable X but you can have as many explanation variables as you want. If you have 2, say, you'll be fitting a plane to the points (x(p), y(p), ln(z(p)), p = 1,...,k. Same goes for any other number of dimensions. For 2 variables you still can sketch a 3D graph.
Just discovered your content. You are a life savior! Thank you
Thanks for posting this! This is helpful added info as I struggle through the MIT Micromasters in Data Science program.
Hello. I would like to ask a question, are there any test statistics to possibly use for underdispersed count data aside from generalized poisson model?
Hi, when you do the scatter plot between Ln(y) and X, which Xp are you considering? If I am not mistaken when you have more than one covariate you can´t graph the multiple linear regression
The graph for simplicity shows one variable X but you can have as many explanation variables as you want. If you have 2, say, you'll be fitting a plane to the points (x(p), y(p), ln(z(p)), p = 1,...,k. Same goes for any other number of dimensions. For 2 variables you still can sketch a 3D graph.