Linear mixed models: how to get adjusted whole model values?

Hi all,

Does anyone know if there is a way of obtaining the "adjusted whole model" (x-axis) values from fitlm?

It is possible to obtain the "fitted" (y-axis) values, but I cannot do this for the x-axis values. I ask this because in the graph below I would like to convert the x-axis values from negative to positive for interpretation purposes.

Any suggestions?

Answer 1

[the cyclist]

I'm not sure if there is a more fundamental way to do this, but you can extract the data from the plot.

load carsmall
Year = categorical(Model_Year);
tbl = table(MPG,Weight,Year);
mdl = fitlm(tbl,'MPG ~ Year + Weight^2');

figure
h = plotAdded(mdl);

% h has three handles. In this case, the first one has the data.
% (In your case, you may need to look at h to figure out which handle is correct.)
xData = h(1).XData;

% Show a few values
xData(1:5)

ans = 1×5

-4.5401 -4.6563 -4.4950 -4.4930 -4.5038