In regression analysis, partial leverage (PL) is a measure of the contribution of the individual independent variables to the total leverage of each observation. That is, if h<sub>i</sub> is the i<sup>th</sup> element of the diagonal of the hat matrix, PL is a measure of how h<sub>i</sub> changes as a variable is added to the regression model. It is computed as:
where
Note that the partial leverage is the leverage of the i<sup>th</sup> point in the partial regression plot for the j<sup>th</sup> variable. Data points with large partial leverage for an independent variable can exert undue influence on the selection of that variable in automatic regression model building procedures.